LMM
2024-08-02
load packages for Linear Mixed Model analysis
3 Result (The following headings respond to the manuscript sections)
3.1 Manipulation checks and results
3.1.1 KSS ratings
# read data
KSSdata <- read_excel("/Users/betty/Desktop/KSS(processed).xlsx", sheet=1)
# as factors
KSSdata$condition <- as.factor(KSSdata$condition)
KSSdata$`agegroup2` <- as.factor(KSSdata$`agegroup2`)
KSSdata$beforeafter <- as.factor(KSSdata$beforeafter)
# groupby
KSSdata %>%
group_by(condition, `agegroup2`, beforeafter)## # A tibble: 290 × 7
## # Groups: condition, agegroup2, beforeafter [20]
## sub age agegroup2 condition order beforeafter score
## <dbl> <dbl> <fct> <fct> <dbl> <fct> <dbl>
## 1 1 42 M control 1 before 7
## 2 2 36 M control 1 before 7
## 3 3 24 Y control 1 before 7
## 4 4 30 Y control 5 before 7
## 5 5 35 Y control 3 before 7
## 6 6 43 M control 3 before 7
## 7 7 26 Y control 4 before 7
## 8 8 24 Y control 5 before 7.5
## 9 9 25 Y control 4 before 6
## 10 10 36 M control 5 before 9
## # ℹ 280 more rows# younger drivers
filter.young<- filter(KSSdata, agegroup2 =="Y")
## main effect of countermeasures for before countermeasure
filter1<- filter(filter.young, beforeafter =="before")
res.fried.KSS.young.before <- filter1 %>% friedman_test(score ~ condition |sub)
res.fried.KSS.young.before## # A tibble: 1 × 6
## .y. n statistic df p method
## * <chr> <int> <dbl> <dbl> <dbl> <chr>
## 1 score 15 3.59 4 0.464 Friedman test## post hoc analysis of countermeasures for before countermeasure
pwc.KSS.inter.young.before <- filter1 %>% wilcox_test(score ~ condition, paired = TRUE, p.adjust.method = "bonferroni")
pwc.KSS.inter.young.before## # A tibble: 10 × 9
## .y. group1 group2 n1 n2 statistic p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <chr>
## 1 score answer control 15 15 27 0.227 1 ns
## 2 score answer meaningfu… 15 15 21 0.525 1 ns
## 3 score answer meaningle… 15 15 33 0.589 1 ns
## 4 score answer repeat 15 15 37 0.904 1 ns
## 5 score control meaningfu… 15 15 9 0.112 1 ns
## 6 score control meaningle… 15 15 14.5 0.66 1 ns
## 7 score control repeat 15 15 14 0.178 1 ns
## 8 score meaningfully meaningle… 15 15 31.5 0.275 1 ns
## 9 score meaningfully repeat 15 15 16.5 0.731 1 ns
## 10 score meaninglessly repeat 15 15 15 0.383 1 ns## main effect of countermeasures for after countermeasure
filter1<- filter(filter.young, beforeafter =="after")
res.fried.KSS.young.after <- filter1 %>% friedman_test(score ~ condition |sub)
res.fried.KSS.young.after## # A tibble: 1 × 6
## .y. n statistic df p method
## * <chr> <int> <dbl> <dbl> <dbl> <chr>
## 1 score 15 29.8 4 0.00000534 Friedman test## post hoc analysis of countermeasures for after countermeasure
pwc.KSS.inter.young.after <- filter1 %>% wilcox_test(score ~ condition, paired = TRUE, p.adjust.method = "bonferroni")
pwc.KSS.inter.young.after## # A tibble: 10 × 9
## .y. group1 group2 n1 n2 statistic p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <chr>
## 1 score answer control 15 15 0 6.98e-4 0.007 **
## 2 score answer meaning… 15 15 7 3 e-3 0.027 *
## 3 score answer meaning… 15 15 7 3 e-3 0.028 *
## 4 score answer repeat 15 15 12.5 4 e-2 0.397 ns
## 5 score control meaning… 15 15 77.5 2.5 e-2 0.249 ns
## 6 score control meaning… 15 15 60.5 8.8 e-2 0.884 ns
## 7 score control repeat 15 15 115 2 e-3 0.017 *
## 8 score meaningfully meaning… 15 15 13.5 3.09e-1 1 ns
## 9 score meaningfully repeat 15 15 69 1.7 e-2 0.167 ns
## 10 score meaninglessly repeat 15 15 80 1.7 e-2 0.167 ns# middle-aged drivers
filter.elder<- filter(KSSdata, agegroup2 =="M")
## main effect of countermeasures for before countermeasure
filter1<- filter(filter.elder, beforeafter =="before")
res.fried.KSS.middle.before <- filter1 %>% friedman_test(score ~ condition |sub)
res.fried.KSS.middle.before## # A tibble: 1 × 6
## .y. n statistic df p method
## * <chr> <int> <dbl> <dbl> <dbl> <chr>
## 1 score 14 10.5 4 0.0330 Friedman test## post hoc analysis of countermeasures for before countermeasure
pwc.KSS.inter.middle.before <- filter1 %>% wilcox_test(score ~ condition, paired = TRUE, p.adjust.method = "bonferroni")
pwc.KSS.inter.middle.before## # A tibble: 10 × 9
## .y. group1 group2 n1 n2 statistic p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <chr>
## 1 score answer control 14 14 60 0.015 0.153 ns
## 2 score answer meaningfu… 14 14 34 0.17 1 ns
## 3 score answer meaningle… 14 14 24 0.105 1 ns
## 4 score answer repeat 14 14 14 1 1 ns
## 5 score control meaningfu… 14 14 5.5 0.079 0.788 ns
## 6 score control meaningle… 14 14 14 0.337 1 ns
## 7 score control repeat 14 14 9 0.033 0.333 ns
## 8 score meaningfully meaningle… 14 14 28.5 0.506 1 ns
## 9 score meaningfully repeat 14 14 9 0.213 1 ns
## 10 score meaninglessly repeat 14 14 7 0.072 0.725 ns## main effect of countermeasures for after countermeasure
filter1<- filter(filter.elder, beforeafter =="after")
res.fried.KSS.middle.after <- filter1 %>% friedman_test(score ~ condition |sub)
res.fried.KSS.middle.after## # A tibble: 1 × 6
## .y. n statistic df p method
## * <chr> <int> <dbl> <dbl> <dbl> <chr>
## 1 score 14 21.8 4 0.000222 Friedman test## post hoc analysis of countermeasures for after countermeasure
pwc.KSS.inter.middle.after <- filter1 %>% wilcox_test(score ~ condition, paired = TRUE, p.adjust.method = "bonferroni")
pwc.KSS.inter.middle.after## # A tibble: 10 × 9
## .y. group1 group2 n1 n2 statistic p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <chr>
## 1 score answer control 14 14 0 0.004 0.037 *
## 2 score answer meaningfu… 14 14 12 0.123 1 ns
## 3 score answer meaningle… 14 14 2 0.004 0.039 *
## 4 score answer repeat 14 14 16.5 0.074 0.742 ns
## 5 score control meaningfu… 14 14 53.5 0.009 0.088 ns
## 6 score control meaningle… 14 14 37.5 0.321 1 ns
## 7 score control repeat 14 14 61 0.014 0.138 ns
## 8 score meaningfully meaningle… 14 14 14.5 0.032 0.317 ns
## 9 score meaningfully repeat 14 14 20 0.809 1 ns
## 10 score meaninglessly repeat 14 14 48.5 0.035 0.35 ns3.1.2 PVT accuracy
# read data
databutton <- read_excel("/Users/betty/Desktop/buttonprocessed.xlsx", sheet=1)
databutton <- select(databutton,"sub","agegroup2","week","condition","order","beforeafter","score2")
# as factors
databutton$condition <- as.factor(databutton$condition)
databutton$beforeafter <- as.factor(databutton$beforeafter)
databutton$agegroup2 <- as.factor(databutton$agegroup2)
# Hypothesis testing
## QQ plot for residual
lmmodel <- lm(score2 ~ condition, databutton)
summary(lmmodel)##
## Call:
## lm(formula = score2 ~ condition, data = databutton)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.50862 -0.15352 -0.00517 0.14436 0.53703
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.56200 0.02877 19.536 <2e-16 ***
## conditioncontrol -0.10848 0.04068 -2.666 0.0081 **
## conditionmeaningfully -0.05683 0.04068 -1.397 0.1636
## conditionmeaninglessly -0.09903 0.04068 -2.434 0.0155 *
## conditionrepeat -0.05338 0.04068 -1.312 0.1906
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2191 on 285 degrees of freedom
## Multiple R-squared: 0.03068, Adjusted R-squared: 0.01708
## F-statistic: 2.255 on 4 and 285 DF, p-value: 0.06334res_lmmodel <- residuals(lmmodel)
qqnorm(res_lmmodel) 
## KS test for residual
ks.test(res_lmmodel, "pnorm", mean(res_lmmodel), sd(res_lmmodel))## Warning in ks.test.default(res_lmmodel, "pnorm", mean(res_lmmodel),
## sd(res_lmmodel)): ties should not be present for the one-sample
## Kolmogorov-Smirnov test##
## Asymptotic one-sample Kolmogorov-Smirnov test
##
## data: res_lmmodel
## D = 0.068894, p-value = 0.1275
## alternative hypothesis: two-sided## Levene’s test for residual
library(car)## Loading required package: carData##
## Attaching package: 'car'## The following object is masked from 'package:dplyr':
##
## recodeleveneTest(res_lmmodel, databutton$condition)## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 4 1.2621 0.2851
## 285# younger drivers
filter.Y <- filter(databutton, agegroup2 =="Y")
## main effect of countermeasures for before countermeasure
filter1.before <- filter(filter.Y, beforeafter =="beforeintervene")
Model <- lmer(data = filter1.before, score2 ~condition +(1|sub) )
summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: score2 ~ condition + (1 | sub)
## Data: filter1.before
##
## REML criterion at convergence: -12.9
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.40298 -0.50155 0.03663 0.61094 2.51597
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.002588 0.05087
## Residual 0.037832 0.19450
## Number of obs: 75, groups: sub, 15
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.482400 0.051910 68.870483 9.293 8.92e-14 ***
## conditioncontrol 0.004267 0.071023 56.000000 0.060 0.952
## conditionmeaningfully -0.069067 0.071023 56.000000 -0.972 0.335
## conditionmeaninglessly -0.093267 0.071023 56.000000 -1.313 0.194
## conditionrepeat -0.075733 0.071023 56.000000 -1.066 0.291
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.684
## cndtnmnngfl -0.684 0.500
## cndtnmnngls -0.684 0.500 0.500
## conditinrpt -0.684 0.500 0.500 0.500anova(Model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 0.12435 0.031088 4 56 0.8217 0.5168## post hoc analysis of countermeasures for before countermeasure
pwc.PVT.inter.young.before <- filter1.before %>% pairwise_t_test(score2 ~ condition, p.adjust.method = "bonferroni", paired = TRUE)
pwc.PVT.inter.young.before ## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 score2 answer contr… 15 15 -0.0652 14 0.949 1 ns
## 2 score2 answer meani… 15 15 1.11 14 0.285 1 ns
## 3 score2 answer meani… 15 15 1.25 14 0.23 1 ns
## 4 score2 answer repeat 15 15 1.86 14 0.084 0.835 ns
## 5 score2 control meani… 15 15 1.05 14 0.313 1 ns
## 6 score2 control meani… 15 15 1.15 14 0.268 1 ns
## 7 score2 control repeat 15 15 0.993 14 0.338 1 ns
## 8 score2 meaningfu… meani… 15 15 0.307 14 0.763 1 ns
## 9 score2 meaningfu… repeat 15 15 0.107 14 0.916 1 ns
## 10 score2 meaningle… repeat 15 15 -0.217 14 0.831 1 ns## main effect of countermeasures for after countermeasure
filter1.after<- filter(filter.Y, beforeafter =="afterintervene")
Model.PVT.2 <- lmer(data = filter1.after, score2~condition +(1|sub) )
summary(Model.PVT.2)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: score2 ~ condition + (1 | sub)
## Data: filter1.after
##
## REML criterion at convergence: -1.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.33457 -0.57873 0.04228 0.55152 1.92542
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.005999 0.07745
## Residual 0.042569 0.20632
## Number of obs: 75, groups: sub, 15
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.57573 0.05690 65.97438 10.118 4.72e-15 ***
## conditioncontrol -0.11573 0.07534 56.00000 -1.536 0.1301
## conditionmeaningfully -0.12907 0.07534 56.00000 -1.713 0.0922 .
## conditionmeaninglessly -0.13647 0.07534 56.00000 -1.811 0.0754 .
## conditionrepeat -0.07573 0.07534 56.00000 -1.005 0.3191
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.662
## cndtnmnngfl -0.662 0.500
## cndtnmnngls -0.662 0.500 0.500
## conditinrpt -0.662 0.500 0.500 0.500anova(Model.PVT.2)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 0.18962 0.047405 4 56 1.1136 0.3593## post hoc analysis of countermeasures for after countermeasure
pwc.PVT.inter.young.after <- filter1.after %>% pairwise_t_test(score2 ~ condition, p.adjust.method = "bonferroni", paired = TRUE)
pwc.PVT.inter.young.after ## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 score2 answer contr… 15 15 1.72 14 0.108 1 ns
## 2 score2 answer meani… 15 15 1.74 14 0.104 1 ns
## 3 score2 answer meani… 15 15 1.65 14 0.121 1 ns
## 4 score2 answer repeat 15 15 0.816 14 0.428 1 ns
## 5 score2 control meani… 15 15 0.225 14 0.825 1 ns
## 6 score2 control meani… 15 15 0.285 14 0.779 1 ns
## 7 score2 control repeat 15 15 -0.705 14 0.492 1 ns
## 8 score2 meaningfu… meani… 15 15 0.119 14 0.907 1 ns
## 9 score2 meaningfu… repeat 15 15 -0.679 14 0.508 1 ns
## 10 score2 meaningle… repeat 15 15 -0.633 14 0.537 1 ns# middle-aged group
filter.M <- filter(databutton, agegroup2 =="M")
## main effect of countermeasures for before countermeasure
filter1.before <- filter(filter.M, beforeafter =="beforeintervene")
Model <- lmer(data = filter1.before, score2 ~condition +(1|sub) )## boundary (singular) fit: see help('isSingular')summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: score2 ~ condition + (1 | sub)
## Data: filter1.before
##
## REML criterion at convergence: -2.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.0509 -0.6224 -0.1950 0.8359 2.3852
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.00000 0.0000
## Residual 0.04601 0.2145
## Number of obs: 70, groups: sub, 14
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.54371 0.05733 65.00000 9.484 7.02e-14 ***
## conditioncontrol -0.10379 0.08108 65.00000 -1.280 0.205
## conditionmeaningfully -0.02943 0.08108 65.00000 -0.363 0.718
## conditionmeaninglessly -0.05536 0.08108 65.00000 -0.683 0.497
## conditionrepeat 0.04200 0.08108 65.00000 0.518 0.606
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.707
## cndtnmnngfl -0.707 0.500
## cndtnmnngls -0.707 0.500 0.500
## conditinrpt -0.707 0.500 0.500 0.500
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')anova(Model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 0.17037 0.042593 4 65 0.9256 0.4546## post hoc analysis of countermeasures for before countermeasure
pwc.PVT.inter.middle.before <- filter1.before %>% pairwise_t_test(score2 ~ condition, p.adjust.method = "bonferroni", paired = TRUE)
pwc.PVT.inter.middle.before## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 score2 answer contr… 14 14 0.998 13 0.336 1 ns
## 2 score2 answer meani… 14 14 0.341 13 0.738 1 ns
## 3 score2 answer meani… 14 14 0.638 13 0.534 1 ns
## 4 score2 answer repeat 14 14 -0.571 13 0.578 1 ns
## 5 score2 control meani… 14 14 -0.863 13 0.404 1 ns
## 6 score2 control meani… 14 14 -0.492 13 0.631 1 ns
## 7 score2 control repeat 14 14 -1.35 13 0.2 1 ns
## 8 score2 meaningfu… meani… 14 14 0.295 13 0.773 1 ns
## 9 score2 meaningfu… repeat 14 14 -0.836 13 0.418 1 ns
## 10 score2 meaningle… repeat 14 14 -2.01 13 0.066 0.655 ns## main effect of countermeasures for after countermeasure
filter1.after<- filter(filter.M, beforeafter =="afterintervene")
Model <- lmer(data = filter1.after, score2~condition +(1|sub) )## boundary (singular) fit: see help('isSingular')summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: score2 ~ condition + (1 | sub)
## Data: filter1.after
##
## REML criterion at convergence: 3.2
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.9334 -0.4811 0.1913 0.6657 2.1219
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.00000 0.000
## Residual 0.05019 0.224
## Number of obs: 70, groups: sub, 14
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.650857 0.059872 65.000000 10.871 2.9e-16 ***
## conditioncontrol -0.226214 0.084672 65.000000 -2.672 0.00953 **
## conditionmeaningfully 0.006286 0.084672 65.000000 0.074 0.94105
## conditionmeaninglessly -0.108786 0.084672 65.000000 -1.285 0.20343
## conditionrepeat -0.100857 0.084672 65.000000 -1.191 0.23793
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.707
## cndtnmnngfl -0.707 0.500
## cndtnmnngls -0.707 0.500 0.500
## conditinrpt -0.707 0.500 0.500 0.500
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')anova(Model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 0.50838 0.12709 4 65 2.5325 0.04861 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## post hoc analysis of countermeasures for after countermeasure
pwc.PVT.inter.middle.after <- filter1.after %>% pairwise_t_test(score2 ~ condition, p.adjust.method = "bonferroni", paired = TRUE)
pwc.PVT.inter.middle.after## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 score2 answer contr… 14 14 2.72 13 0.018 0.175 ns
## 2 score2 answer meani… 14 14 -0.0777 13 0.939 1 ns
## 3 score2 answer meani… 14 14 1.62 13 0.128 1 ns
## 4 score2 answer repeat 14 14 1.34 13 0.203 1 ns
## 5 score2 control meani… 14 14 -2.44 13 0.03 0.296 ns
## 6 score2 control meani… 14 14 -1.18 13 0.258 1 ns
## 7 score2 control repeat 14 14 -1.12 13 0.281 1 ns
## 8 score2 meaningfu… meani… 14 14 1.25 13 0.233 1 ns
## 9 score2 meaningfu… repeat 14 14 1.55 13 0.145 1 ns
## 10 score2 meaningle… repeat 14 14 -0.101 13 0.921 1 ns3.2 Driving performance
3.2.1 SDLP
##
## Call:
## lm(formula = SDLP ~ condition * beforeafter, data = SDLPdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.88576 -0.25101 -0.04114 0.18767 1.29628
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.01759 0.07000 14.537 <2e-16 ***
## conditioncontrol -0.12548 0.09900 -1.268 0.2060
## conditionmeaningfully -0.03852 0.09900 -0.389 0.6975
## conditionmeaninglessly -0.09040 0.09900 -0.913 0.3619
## conditionrepeat 0.13517 0.09900 1.365 0.1732
## beforeafteron -0.18586 0.09900 -1.877 0.0615 .
## conditioncontrol:beforeafteron 0.36069 0.14000 2.576 0.0105 *
## conditionmeaningfully:beforeafteron 0.22524 0.14000 1.609 0.1088
## conditionmeaninglessly:beforeafteron 0.34530 0.14000 2.466 0.0142 *
## conditionrepeat:beforeafteron -0.17052 0.14000 -1.218 0.2243
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.377 on 280 degrees of freedom
## Multiple R-squared: 0.07906, Adjusted R-squared: 0.04946
## F-statistic: 2.671 on 9 and 280 DF, p-value: 0.005463
## Warning in ks.test.default(res_lmmodel, "pnorm", mean(res_lmmodel),
## sd(res_lmmodel)): ties should not be present for the one-sample
## Kolmogorov-Smirnov test##
## Asymptotic one-sample Kolmogorov-Smirnov test
##
## data: res_lmmodel
## D = 0.074915, p-value = 0.07715
## alternative hypothesis: two-sided## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 9 1.8339 0.06212 .
## 280
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Call:
## lm(formula = log(SDLP) ~ condition * beforeafter, data = SDLPdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.36356 -0.23705 0.01497 0.26650 1.00415
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.07155 0.07178 -0.997 0.3197
## conditioncontrol -0.12347 0.10152 -1.216 0.2249
## conditionmeaningfully -0.04054 0.10152 -0.399 0.6899
## conditionmeaninglessly -0.04567 0.10152 -0.450 0.6532
## conditionrepeat 0.11460 0.10152 1.129 0.2599
## beforeafteron -0.17742 0.10152 -1.748 0.0816 .
## conditioncontrol:beforeafteron 0.34692 0.14357 2.416 0.0163 *
## conditionmeaningfully:beforeafteron 0.25314 0.14357 1.763 0.0790 .
## conditionmeaninglessly:beforeafteron 0.33215 0.14357 2.314 0.0214 *
## conditionrepeat:beforeafteron -0.13153 0.14357 -0.916 0.3604
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3866 on 280 degrees of freedom
## Multiple R-squared: 0.07021, Adjusted R-squared: 0.04033
## F-statistic: 2.349 on 9 and 280 DF, p-value: 0.01436
## Warning in ks.test.default(res_lmmodel, "pnorm", mean(res_lmmodel),
## sd(res_lmmodel)): ties should not be present for the one-sample
## Kolmogorov-Smirnov test##
## Asymptotic one-sample Kolmogorov-Smirnov test
##
## data: res_lmmodel
## D = 0.054475, p-value = 0.3557
## alternative hypothesis: two-sided## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 9 1.1252 0.3447
## 280## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: SDLP ~ condition * beforeafter * agegroup2 + (1 | sub)
## Data: SDLPdata
##
## REML criterion at convergence: 238.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -4.0071 -0.5049 -0.0318 0.5780 2.9069
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.05821 0.2413
## Residual 0.09549 0.3090
## Number of obs: 290, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error
## (Intercept) -0.075832 0.104776
## conditioncontrol -0.099603 0.116795
## conditionmeaningfully -0.050611 0.116795
## conditionmeaninglessly -0.021721 0.116795
## conditionrepeat 0.120110 0.116795
## beforeafteron -0.125023 0.116795
## agegroup2Y 0.008273 0.145685
## conditioncontrol:beforeafteron 0.335424 0.165173
## conditionmeaningfully:beforeafteron 0.157296 0.165173
## conditionmeaninglessly:beforeafteron 0.215010 0.165173
## conditionrepeat:beforeafteron -0.164461 0.165173
## conditioncontrol:agegroup2Y -0.046144 0.162396
## conditionmeaningfully:agegroup2Y 0.019464 0.162396
## conditionmeaninglessly:agegroup2Y -0.046295 0.162396
## conditionrepeat:agegroup2Y -0.010645 0.162396
## beforeafteron:agegroup2Y -0.101294 0.162396
## conditioncontrol:beforeafteron:agegroup2Y 0.022221 0.229663
## conditionmeaningfully:beforeafteron:agegroup2Y 0.185299 0.229663
## conditionmeaninglessly:beforeafteron:agegroup2Y 0.226475 0.229663
## conditionrepeat:beforeafteron:agegroup2Y 0.063661 0.229663
## df t value Pr(>|t|)
## (Intercept) 117.862084 -0.724 0.4706
## conditioncontrol 243.000000 -0.853 0.3946
## conditionmeaningfully 243.000000 -0.433 0.6652
## conditionmeaninglessly 243.000000 -0.186 0.8526
## conditionrepeat 243.000000 1.028 0.3048
## beforeafteron 243.000000 -1.070 0.2855
## agegroup2Y 117.862084 0.057 0.9548
## conditioncontrol:beforeafteron 243.000000 2.031 0.0434 *
## conditionmeaningfully:beforeafteron 243.000000 0.952 0.3419
## conditionmeaninglessly:beforeafteron 243.000000 1.302 0.1942
## conditionrepeat:beforeafteron 243.000000 -0.996 0.3204
## conditioncontrol:agegroup2Y 243.000000 -0.284 0.7765
## conditionmeaningfully:agegroup2Y 243.000000 0.120 0.9047
## conditionmeaninglessly:agegroup2Y 243.000000 -0.285 0.7758
## conditionrepeat:agegroup2Y 243.000000 -0.066 0.9478
## beforeafteron:agegroup2Y 243.000000 -0.624 0.5334
## conditioncontrol:beforeafteron:agegroup2Y 243.000000 0.097 0.9230
## conditionmeaningfully:beforeafteron:agegroup2Y 243.000000 0.807 0.4206
## conditionmeaninglessly:beforeafteron:agegroup2Y 243.000000 0.986 0.3251
## conditionrepeat:beforeafteron:agegroup2Y 243.000000 0.277 0.7819
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Correlation matrix not shown by default, as p = 20 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 0.45834 0.11459 4 243 1.2000 0.3114
## beforeafter 0.02155 0.02155 1 243 0.2257 0.6352
## agegroup2 0.00089 0.00089 1 27 0.0093 0.9238
## condition:beforeafter 2.64184 0.66046 4 243 6.9168 2.742e-05
## condition:agegroup2 0.19216 0.04804 4 243 0.5031 0.7335
## beforeafter:agegroup2 0.00006 0.00006 1 243 0.0006 0.9806
## condition:beforeafter:agegroup2 0.14715 0.03679 4 243 0.3853 0.8191
##
## condition
## beforeafter
## agegroup2
## condition:beforeafter ***
## condition:agegroup2
## beforeafter:agegroup2
## condition:beforeafter:agegroup2
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: SDLP ~ condition + (1 | sub)
## Data: filter1.before
##
## REML criterion at convergence: 148.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.5595 -0.4646 -0.0047 0.7476 1.9585
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.0634 0.2518
## Residual 0.1152 0.3394
## Number of obs: 145, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.07155 0.07848 93.09034 -0.912 0.364
## conditioncontrol -0.12347 0.08914 112.00000 -1.385 0.169
## conditionmeaningfully -0.04054 0.08914 112.00000 -0.455 0.650
## conditionmeaninglessly -0.04567 0.08914 112.00000 -0.512 0.609
## conditionrepeat 0.11460 0.08914 112.00000 1.286 0.201
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.568
## cndtnmnngfl -0.568 0.500
## cndtnmnngls -0.568 0.500 0.500
## conditinrpt -0.568 0.500 0.500 0.500## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 0.87871 0.21968 4 112 1.9065 0.1143## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 SDLP answer contr… 29 29 1.37 28 0.183 1 ns
## 2 SDLP answer meani… 29 29 0.411 28 0.684 1 ns
## 3 SDLP answer meani… 29 29 0.480 28 0.635 1 ns
## 4 SDLP answer repeat 29 29 -1.53 28 0.137 1 ns
## 5 SDLP control meani… 29 29 -0.842 28 0.407 1 ns
## 6 SDLP control meani… 29 29 -1.12 28 0.273 1 ns
## 7 SDLP control repeat 29 29 -2.85 28 0.008 0.082 ns
## 8 SDLP meaningful… meani… 29 29 0.0526 28 0.958 1 ns
## 9 SDLP meaningful… repeat 29 29 -1.74 28 0.092 0.924 ns
## 10 SDLP meaningles… repeat 29 29 -1.80 28 0.082 0.821 ns## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: SDLP ~ condition + (1 | sub)
## Data: filter1.on
##
## REML criterion at convergence: 92.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.47044 -0.51527 -0.03246 0.50840 3.12998
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.04311 0.2076
## Residual 0.07714 0.2777
## Number of obs: 145, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.24897 0.06439 92.46116 -3.866 0.000205 ***
## conditioncontrol 0.22345 0.07294 112.00000 3.064 0.002741 **
## conditionmeaningfully 0.21260 0.07294 112.00000 2.915 0.004300 **
## conditionmeaninglessly 0.28648 0.07294 112.00000 3.928 0.000149 ***
## conditionrepeat -0.01693 0.07294 112.00000 -0.232 0.816873
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.566
## cndtnmnngfl -0.566 0.500
## cndtnmnngls -0.566 0.500 0.500
## conditinrpt -0.566 0.500 0.500 0.500## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 2.2595 0.56486 4 112 7.3226 2.822e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 SDLP answer contr… 29 29 -2.32 28 2.8 e-2 2.77e-1 ns
## 2 SDLP answer meani… 29 29 -3.00 28 6 e-3 5.6 e-2 ns
## 3 SDLP answer meani… 29 29 -4.14 28 2.86e-4 3 e-3 **
## 4 SDLP answer repeat 29 29 0.282 28 7.8 e-1 1 e+0 ns
## 5 SDLP control meani… 29 29 0.138 28 8.91e-1 1 e+0 ns
## 6 SDLP control meani… 29 29 -0.739 28 4.66e-1 1 e+0 ns
## 7 SDLP control repeat 29 29 3.32 28 3 e-3 2.5 e-2 *
## 8 SDLP meanin… meani… 29 29 -1.20 28 2.4 e-1 1 e+0 ns
## 9 SDLP meanin… repeat 29 29 3.78 28 7.54e-4 8 e-3 **
## 10 SDLP meanin… repeat 29 29 4.57 28 8.91e-5 8.91e-4 ***3.2.2 SDVHA
##
## Call:
## lm(formula = SDVH ~ condition * beforeafter, data = SDVHdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.67721 -0.23507 -0.06415 0.12161 2.25838
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.75321 0.07107 10.598 <2e-16 ***
## conditioncontrol -0.11286 0.10051 -1.123 0.2624
## conditionmeaningfully -0.05100 0.10051 -0.507 0.6123
## conditionmeaninglessly -0.10562 0.10051 -1.051 0.2942
## conditionrepeat 0.10641 0.10051 1.059 0.2906
## beforeafteron -0.19048 0.10051 -1.895 0.0591 .
## conditioncontrol:beforeafteron 0.30479 0.14214 2.144 0.0329 *
## conditionmeaningfully:beforeafteron 0.16610 0.14214 1.169 0.2436
## conditionmeaninglessly:beforeafteron 0.37070 0.14214 2.608 0.0096 **
## conditionrepeat:beforeafteron -0.14762 0.14214 -1.039 0.2999
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3827 on 280 degrees of freedom
## Multiple R-squared: 0.06902, Adjusted R-squared: 0.0391
## F-statistic: 2.307 on 9 and 280 DF, p-value: 0.0163
## Warning in ks.test.default(res_lmmodel, "pnorm", mean(res_lmmodel),
## sd(res_lmmodel)): ties should not be present for the one-sample
## Kolmogorov-Smirnov test##
## Asymptotic one-sample Kolmogorov-Smirnov test
##
## data: res_lmmodel
## D = 0.1398, p-value = 2.387e-05
## alternative hypothesis: two-sided## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 9 0.6792 0.7276
## 280##
## Call:
## lm(formula = log(SDVH) ~ condition * beforeafter, data = SDVHdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.1756 -0.3213 0.0310 0.2939 1.4348
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.40733 0.09527 -4.275 2.62e-05 ***
## conditioncontrol -0.14713 0.13474 -1.092 0.2758
## conditionmeaningfully -0.10595 0.13474 -0.786 0.4323
## conditionmeaninglessly -0.11304 0.13474 -0.839 0.4022
## conditionrepeat 0.10968 0.13474 0.814 0.4163
## beforeafteron -0.27484 0.13474 -2.040 0.0423 *
## conditioncontrol:beforeafteron 0.42540 0.19055 2.233 0.0264 *
## conditionmeaningfully:beforeafteron 0.28109 0.19055 1.475 0.1413
## conditionmeaninglessly:beforeafteron 0.48545 0.19055 2.548 0.0114 *
## conditionrepeat:beforeafteron -0.19366 0.19055 -1.016 0.3104
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.5131 on 280 degrees of freedom
## Multiple R-squared: 0.07378, Adjusted R-squared: 0.044
## F-statistic: 2.478 on 9 and 280 DF, p-value: 0.009795
## Warning in ks.test.default(res_lmmodel, "pnorm", mean(res_lmmodel),
## sd(res_lmmodel)): ties should not be present for the one-sample
## Kolmogorov-Smirnov test##
## Asymptotic one-sample Kolmogorov-Smirnov test
##
## data: res_lmmodel
## D = 0.04506, p-value = 0.5981
## alternative hypothesis: two-sided## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 9 0.208 0.9931
## 280## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: SDVH ~ condition * beforeafter * agegroup2 + (1 | sub)
## Data: SDVHdata
##
## REML criterion at convergence: 363.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -7.5611 -0.4774 0.0161 0.5390 2.0382
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.1181 0.3436
## Residual 0.1482 0.3850
## Number of obs: 290, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error
## (Intercept) -0.503660 0.137914
## conditioncontrol 0.004855 0.145507
## conditionmeaningfully -0.174970 0.145507
## conditionmeaninglessly -0.055690 0.145507
## conditionrepeat 0.112300 0.145507
## beforeafteron -0.148515 0.145507
## agegroup2Y 0.186229 0.191762
## conditioncontrol:beforeafteron 0.269060 0.205778
## conditionmeaningfully:beforeafteron 0.217282 0.205778
## conditionmeaninglessly:beforeafteron 0.332222 0.205778
## conditionrepeat:beforeafteron -0.245187 0.205778
## conditioncontrol:agegroup2Y -0.293846 0.202319
## conditionmeaningfully:agegroup2Y 0.133443 0.202319
## conditionmeaninglessly:agegroup2Y -0.110883 0.202319
## conditionrepeat:agegroup2Y -0.005061 0.202319
## beforeafteron:agegroup2Y -0.244223 0.202319
## conditioncontrol:beforeafteron:agegroup2Y 0.302254 0.286122
## conditionmeaningfully:beforeafteron:agegroup2Y 0.123362 0.286122
## conditionmeaninglessly:beforeafteron:agegroup2Y 0.296247 0.286122
## conditionrepeat:beforeafteron:agegroup2Y 0.099622 0.286122
## df t value Pr(>|t|)
## (Intercept) 97.483775 -3.652 0.000421 ***
## conditioncontrol 243.000000 0.033 0.973412
## conditionmeaningfully 243.000000 -1.202 0.230345
## conditionmeaninglessly 243.000000 -0.383 0.702251
## conditionrepeat 243.000000 0.772 0.440993
## beforeafteron 243.000000 -1.021 0.308423
## agegroup2Y 97.483775 0.971 0.333879
## conditioncontrol:beforeafteron 243.000000 1.308 0.192269
## conditionmeaningfully:beforeafteron 243.000000 1.056 0.292060
## conditionmeaninglessly:beforeafteron 243.000000 1.614 0.107723
## conditionrepeat:beforeafteron 243.000000 -1.192 0.234615
## conditioncontrol:agegroup2Y 243.000000 -1.452 0.147683
## conditionmeaningfully:agegroup2Y 243.000000 0.660 0.510155
## conditionmeaninglessly:agegroup2Y 243.000000 -0.548 0.584153
## conditionrepeat:agegroup2Y 243.000000 -0.025 0.980064
## beforeafteron:agegroup2Y 243.000000 -1.207 0.228559
## conditioncontrol:beforeafteron:agegroup2Y 243.000000 1.056 0.291844
## conditionmeaningfully:beforeafteron:agegroup2Y 243.000000 0.431 0.666739
## conditionmeaninglessly:beforeafteron:agegroup2Y 243.000000 1.035 0.301518
## conditionrepeat:beforeafteron:agegroup2Y 243.000000 0.348 0.728006
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Correlation matrix not shown by default, as p = 20 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 0.6282 0.15705 4 243 1.0597 0.3771
## beforeafter 0.3944 0.39443 1 243 2.6614 0.1041
## agegroup2 0.0669 0.06687 1 27 0.4512 0.5075
## condition:beforeafter 4.7733 1.19332 4 243 8.0518 4.12e-06 ***
## condition:agegroup2 0.8432 0.21079 4 243 1.4223 0.2271
## beforeafter:agegroup2 0.1156 0.11565 1 243 0.7803 0.3779
## condition:beforeafter:agegroup2 0.2509 0.06272 4 243 0.4232 0.7918
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: SDVH ~ condition + (1 | sub)
## Data: filter1.before
##
## REML criterion at convergence: 222.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -6.3529 -0.4123 0.0272 0.5219 2.0906
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.0968 0.3111
## Residual 0.1993 0.4464
## Number of obs: 145, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.4073 0.1010 98.0678 -4.031 0.00011 ***
## conditioncontrol -0.1471 0.1172 112.0000 -1.255 0.21206
## conditionmeaningfully -0.1059 0.1172 112.0000 -0.904 0.36807
## conditionmeaninglessly -0.1130 0.1172 112.0000 -0.964 0.33698
## conditionrepeat 0.1097 0.1172 112.0000 0.936 0.35149
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.580
## cndtnmnngfl -0.580 0.500
## cndtnmnngls -0.580 0.500 0.500
## conditinrpt -0.580 0.500 0.500 0.500## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 1.2914 0.32284 4 112 1.6201 0.1741## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 SDVH answer contr… 29 29 1.39 28 0.175 1 ns
## 2 SDVH answer meani… 29 29 0.801 28 0.43 1 ns
## 3 SDVH answer meani… 29 29 1.06 28 0.298 1 ns
## 4 SDVH answer repeat 29 29 -1.27 28 0.215 1 ns
## 5 SDVH control meani… 29 29 -0.287 28 0.776 1 ns
## 6 SDVH control meani… 29 29 -0.401 28 0.692 1 ns
## 7 SDVH control repeat 29 29 -3.10 28 0.004 0.044 *
## 8 SDVH meaningful… meani… 29 29 0.0459 28 0.964 1 ns
## 9 SDVH meaningful… repeat 29 29 -1.47 28 0.153 1 ns
## 10 SDVH meaningles… repeat 29 29 -2.25 28 0.032 0.322 ns## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: SDVH ~ condition + (1 | sub)
## Data: filter1.on
##
## REML criterion at convergence: 145.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.34750 -0.53643 -0.00654 0.66065 1.94659
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.13334 0.3652
## Residual 0.09705 0.3115
## Number of obs: 145, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.68217 0.08913 59.83400 -7.653 1.95e-10 ***
## conditioncontrol 0.27826 0.08181 112.00000 3.401 0.000931 ***
## conditionmeaningfully 0.17514 0.08181 112.00000 2.141 0.034462 *
## conditionmeaninglessly 0.37241 0.08181 112.00000 4.552 1.36e-05 ***
## conditionrepeat -0.08398 0.08181 112.00000 -1.026 0.306897
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.459
## cndtnmnngfl -0.459 0.500
## cndtnmnngls -0.459 0.500 0.500
## conditinrpt -0.459 0.500 0.500 0.500## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 4.1696 1.0424 4 112 10.74 2.098e-07 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 SDVH answer contr… 29 29 -3.19 28 3 e-3 3.5 e-2 *
## 2 SDVH answer meani… 29 29 -2.35 28 2.6 e-2 2.63e-1 ns
## 3 SDVH answer meani… 29 29 -4.28 28 1.97e-4 2 e-3 **
## 4 SDVH answer repeat 29 29 1.28 28 2.11e-1 1 e+0 ns
## 5 SDVH control meani… 29 29 1.25 28 2.21e-1 1 e+0 ns
## 6 SDVH control meani… 29 29 -1.16 28 2.57e-1 1 e+0 ns
## 7 SDVH control repeat 29 29 3.89 28 5.7 e-4 6 e-3 **
## 8 SDVH meanin… meani… 29 29 -2.73 28 1.1 e-2 1.09e-1 ns
## 9 SDVH meanin… repeat 29 29 3.43 28 2 e-3 1.9 e-2 *
## 10 SDVH meanin… repeat 29 29 4.85 28 4.19e-5 4.19e-4 ***3.2.3 SDspeed
##
## Call:
## lm(formula = SDSpeed ~ condition * beforeafter, data = SDSpeeddata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.4694 -2.2895 -0.5463 1.0054 18.6664
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.23881 0.64181 5.046 8.11e-07 ***
## conditioncontrol 0.71306 0.90765 0.786 0.4328
## conditionmeaningfully 0.23972 0.90765 0.264 0.7919
## conditionmeaninglessly 0.81206 0.90765 0.895 0.3717
## conditionrepeat -0.31646 0.90765 -0.349 0.7276
## beforeafteron 1.77301 0.90765 1.953 0.0518 .
## conditioncontrol:beforeafteron -1.60608 1.28361 -1.251 0.2119
## conditionmeaningfully:beforeafteron -0.05315 1.28361 -0.041 0.9670
## conditionmeaninglessly:beforeafteron -0.94991 1.28361 -0.740 0.4599
## conditionrepeat:beforeafteron 0.58784 1.28361 0.458 0.6473
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 3.456 on 280 degrees of freedom
## Multiple R-squared: 0.05295, Adjusted R-squared: 0.02251
## F-statistic: 1.74 on 9 and 280 DF, p-value: 0.07996
## Warning in ks.test.default(res_lmmodel, "pnorm", mean(res_lmmodel),
## sd(res_lmmodel)): ties should not be present for the one-sample
## Kolmogorov-Smirnov test##
## Asymptotic one-sample Kolmogorov-Smirnov test
##
## data: res_lmmodel
## D = 0.15155, p-value = 3.28e-06
## alternative hypothesis: two-sided## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 9 0.3938 0.9376
## 280##
## Call:
## lm(formula = log(SDSpeed) ~ condition * beforeafter, data = SDSpeeddata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.2608 -0.4923 0.1509 0.5566 2.3272
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.62800 0.17620 3.564 0.000429 ***
## conditioncontrol 0.16355 0.24918 0.656 0.512131
## conditionmeaningfully 0.16580 0.24918 0.665 0.506341
## conditionmeaninglessly 0.49093 0.24918 1.970 0.049801 *
## conditionrepeat 0.13784 0.24918 0.553 0.580581
## beforeafteron 0.75331 0.24918 3.023 0.002733 **
## conditioncontrol:beforeafteron -0.51673 0.35239 -1.466 0.143670
## conditionmeaningfully:beforeafteron -0.07457 0.35239 -0.212 0.832561
## conditionmeaninglessly:beforeafteron -0.51447 0.35239 -1.460 0.145428
## conditionrepeat:beforeafteron -0.06010 0.35239 -0.171 0.864704
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.9488 on 280 degrees of freedom
## Multiple R-squared: 0.0974, Adjusted R-squared: 0.06839
## F-statistic: 3.357 on 9 and 280 DF, p-value: 0.000636
## Warning in ks.test.default(res_lmmodel, "pnorm", mean(res_lmmodel),
## sd(res_lmmodel)): ties should not be present for the one-sample
## Kolmogorov-Smirnov test##
## Asymptotic one-sample Kolmogorov-Smirnov test
##
## data: res_lmmodel
## D = 0.071556, p-value = 0.1026
## alternative hypothesis: two-sided## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 9 2.1206 0.02794 *
## 280
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: SDSpeed ~ condition * beforeafter * agegroup2 + (1 | sub)
## Data: SDSpeeddata
##
## REML criterion at convergence: 758.7
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -4.5088 -0.4548 0.0742 0.5761 2.4083
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.1737 0.4168
## Residual 0.7046 0.8394
## Number of obs: 290, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df
## (Intercept) 0.55884 0.25047 199.70689
## conditioncontrol 0.81102 0.31727 243.00000
## conditionmeaningfully 0.18126 0.31727 243.00000
## conditionmeaninglessly 0.66972 0.31727 243.00000
## conditionrepeat 0.05963 0.31727 243.00000
## beforeafteron 0.94604 0.31727 243.00000
## agegroup2Y 0.13372 0.34827 199.70689
## conditioncontrol:beforeafteron -0.93847 0.44869 243.00000
## conditionmeaningfully:beforeafteron -0.25804 0.44869 243.00000
## conditionmeaninglessly:beforeafteron -0.93085 0.44869 243.00000
## conditionrepeat:beforeafteron -0.25955 0.44869 243.00000
## conditioncontrol:agegroup2Y -1.25178 0.44115 243.00000
## conditionmeaningfully:agegroup2Y -0.02988 0.44115 243.00000
## conditionmeaninglessly:agegroup2Y -0.34566 0.44115 243.00000
## conditionrepeat:agegroup2Y 0.15121 0.44115 243.00000
## beforeafteron:agegroup2Y -0.37262 0.44115 243.00000
## conditioncontrol:beforeafteron:agegroup2Y 0.81535 0.62388 243.00000
## conditionmeaningfully:beforeafteron:agegroup2Y 0.35470 0.62388 243.00000
## conditionmeaninglessly:beforeafteron:agegroup2Y 0.80501 0.62388 243.00000
## conditionrepeat:beforeafteron:agegroup2Y 0.38560 0.62388 243.00000
## t value Pr(>|t|)
## (Intercept) 2.231 0.02679 *
## conditioncontrol 2.556 0.01119 *
## conditionmeaningfully 0.571 0.56832
## conditionmeaninglessly 2.111 0.03581 *
## conditionrepeat 0.188 0.85108
## beforeafteron 2.982 0.00316 **
## agegroup2Y 0.384 0.70142
## conditioncontrol:beforeafteron -2.092 0.03751 *
## conditionmeaningfully:beforeafteron -0.575 0.56576
## conditionmeaninglessly:beforeafteron -2.075 0.03908 *
## conditionrepeat:beforeafteron -0.578 0.56349
## conditioncontrol:agegroup2Y -2.838 0.00493 **
## conditionmeaningfully:agegroup2Y -0.068 0.94605
## conditionmeaninglessly:agegroup2Y -0.784 0.43407
## conditionrepeat:agegroup2Y 0.343 0.73208
## beforeafteron:agegroup2Y -0.845 0.39914
## conditioncontrol:beforeafteron:agegroup2Y 1.307 0.19248
## conditionmeaningfully:beforeafteron:agegroup2Y 0.569 0.57019
## conditionmeaninglessly:beforeafteron:agegroup2Y 1.290 0.19816
## conditionrepeat:beforeafteron:agegroup2Y 0.618 0.53711
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Correlation matrix not shown by default, as p = 20 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 3.3554 0.8388 4 243 1.1905 0.315560
## beforeafter 19.4620 19.4620 1 243 27.6204 3.23e-07
## agegroup2 0.2609 0.2609 1 27 0.3703 0.547908
## condition:beforeafter 4.0652 1.0163 4 243 1.4423 0.220626
## condition:agegroup2 12.1413 3.0353 4 243 4.3077 0.002195
## beforeafter:agegroup2 0.1793 0.1793 1 243 0.2545 0.614413
## condition:beforeafter:agegroup2 1.7119 0.4280 4 243 0.6074 0.657699
##
## condition
## beforeafter ***
## agegroup2
## condition:beforeafter
## condition:agegroup2 **
## beforeafter:agegroup2
## condition:beforeafter:agegroup2
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## # A tibble: 1 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 SDSpeed befor… on 145 145 -6.15 144 7.16e-9 7.16e-9 ****## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: SDSpeed ~ condition + (1 | sub)
## Data: filter.before
##
## REML criterion at convergence: 435.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.9397 -0.3669 0.0825 0.6628 1.8641
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.1585 0.3981
## Residual 1.0428 1.0212
## Number of obs: 145, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.6280 0.2035 130.8906 3.086 0.00248 **
## conditioncontrol 0.1636 0.2682 112.0000 0.610 0.54319
## conditionmeaningfully 0.1658 0.2682 112.0000 0.618 0.53766
## conditionmeaninglessly 0.4909 0.2682 112.0000 1.831 0.06982 .
## conditionrepeat 0.1378 0.2682 112.0000 0.514 0.60827
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.659
## cndtnmnngfl -0.659 0.500
## cndtnmnngls -0.659 0.500 0.500
## conditinrpt -0.659 0.500 0.500 0.500## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 3.7889 0.94723 4 112 0.9083 0.4618## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 SDSpeed answer contr… 29 29 -0.524 28 0.605 1 ns
## 2 SDSpeed answer meani… 29 29 -0.536 28 0.596 1 ns
## 3 SDSpeed answer meani… 29 29 -2.22 28 0.034 0.344 ns
## 4 SDSpeed answer repeat 29 29 -0.492 28 0.627 1 ns
## 5 SDSpeed control meani… 29 29 -0.00703 28 0.994 1 ns
## 6 SDSpeed control meani… 29 29 -1.47 28 0.152 1 ns
## 7 SDSpeed control repeat 29 29 0.0905 28 0.929 1 ns
## 8 SDSpeed meaningf… meani… 29 29 -1.39 28 0.175 1 ns
## 9 SDSpeed meaningf… repeat 29 29 0.107 28 0.915 1 ns
## 10 SDSpeed meaningl… repeat 29 29 1.70 28 0.1 1 ns## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: SDSpeed ~ condition + (1 | sub)
## Data: filter.during
##
## REML criterion at convergence: 332.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -4.9679 -0.4839 0.1730 0.5953 2.0414
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.1314 0.3626
## Residual 0.4679 0.6840
## Number of obs: 145, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 1.38131 0.14376 117.40883 9.609 <2e-16 ***
## conditioncontrol -0.35318 0.17963 112.00000 -1.966 0.0518 .
## conditionmeaningfully 0.09123 0.17963 112.00000 0.508 0.6125
## conditionmeaninglessly -0.02354 0.17963 112.00000 -0.131 0.8960
## conditionrepeat 0.07774 0.17963 112.00000 0.433 0.6660
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.625
## cndtnmnngfl -0.625 0.500
## cndtnmnngls -0.625 0.500 0.500
## conditinrpt -0.625 0.500 0.500 0.500## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 3.7998 0.94995 4 112 2.0304 0.09493 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 SDSpeed answer contr… 29 29 1.67 28 0.105 1 ns
## 2 SDSpeed answer meani… 29 29 -0.632 28 0.532 1 ns
## 3 SDSpeed answer meani… 29 29 0.180 28 0.858 1 ns
## 4 SDSpeed answer repeat 29 29 -0.599 28 0.554 1 ns
## 5 SDSpeed control meani… 29 29 -1.77 28 0.088 0.876 ns
## 6 SDSpeed control meani… 29 29 -1.39 28 0.175 1 ns
## 7 SDSpeed control repeat 29 29 -1.90 28 0.067 0.673 ns
## 8 SDSpeed meaningf… meani… 29 29 0.779 28 0.442 1 ns
## 9 SDSpeed meaningf… repeat 29 29 0.106 28 0.916 1 ns
## 10 SDSpeed meaningl… repeat 29 29 -0.823 28 0.418 1 ns3.3 Physiological indicators
3.3.1 SCL (skin conductance level extracted from electrodermal activity (EDA))
##
## Call:
## lm(formula = SCL ~ condition * beforeafter, data = SCLdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.2870 -0.3534 -0.0619 0.1784 3.1828
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.16091 0.11472 10.119 < 2e-16 ***
## conditioncontrol -0.06518 0.16225 -0.402 0.688196
## conditionmeaningfully 0.10921 0.16225 0.673 0.501444
## conditionmeaninglessly -0.13121 0.16225 -0.809 0.419368
## conditionrepeat -0.07328 0.16225 -0.452 0.651884
## beforeafteron 0.82918 0.16225 5.111 5.95e-07 ***
## conditioncontrol:beforeafteron -0.90331 0.22945 -3.937 0.000104 ***
## conditionmeaningfully:beforeafteron -0.45642 0.22945 -1.989 0.047653 *
## conditionmeaninglessly:beforeafteron -0.71217 0.22945 -3.104 0.002106 **
## conditionrepeat:beforeafteron -0.11845 0.22945 -0.516 0.606109
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.6178 on 280 degrees of freedom
## Multiple R-squared: 0.2327, Adjusted R-squared: 0.208
## F-statistic: 9.433 on 9 and 280 DF, p-value: 1.543e-12
## Warning in ks.test.default(res_lmmodel, "pnorm", mean(res_lmmodel),
## sd(res_lmmodel)): ties should not be present for the one-sample
## Kolmogorov-Smirnov test##
## Asymptotic one-sample Kolmogorov-Smirnov test
##
## data: res_lmmodel
## D = 0.16347, p-value = 3.711e-07
## alternative hypothesis: two-sided## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 9 4.3732 2.344e-05 ***
## 280
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: SCL ~ condition * beforeafter * agegroup2 + (1 | sub)
## Data: SCLdata
##
## REML criterion at convergence: 557.7
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.9498 -0.5555 -0.0550 0.2955 4.9685
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.02987 0.1728
## Residual 0.35657 0.5971
## Number of obs: 290, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df
## (Intercept) 1.24679 0.16614 256.22060
## conditioncontrol -0.27489 0.22570 243.00000
## conditionmeaningfully 0.15149 0.22570 243.00000
## conditionmeaninglessly -0.18256 0.22570 243.00000
## conditionrepeat -0.23172 0.22570 243.00000
## beforeafteron 0.84353 0.22570 243.00000
## agegroup2Y -0.16604 0.23101 256.22060
## conditioncontrol:beforeafteron -0.84058 0.31918 243.00000
## conditionmeaningfully:beforeafteron -0.51503 0.31918 243.00000
## conditionmeaninglessly:beforeafteron -0.74198 0.31918 243.00000
## conditionrepeat:beforeafteron -0.22876 0.31918 243.00000
## conditioncontrol:agegroup2Y 0.40545 0.31382 243.00000
## conditionmeaningfully:agegroup2Y -0.08175 0.31382 243.00000
## conditionmeaninglessly:agegroup2Y 0.09928 0.31382 243.00000
## conditionrepeat:agegroup2Y 0.30633 0.31382 243.00000
## beforeafteron:agegroup2Y -0.02774 0.31382 243.00000
## conditioncontrol:beforeafteron:agegroup2Y -0.12130 0.44380 243.00000
## conditionmeaningfully:beforeafteron:agegroup2Y 0.11332 0.44380 243.00000
## conditionmeaninglessly:beforeafteron:agegroup2Y 0.05762 0.44380 243.00000
## conditionrepeat:beforeafteron:agegroup2Y 0.21327 0.44380 243.00000
## t value Pr(>|t|)
## (Intercept) 7.504 1.02e-12 ***
## conditioncontrol -1.218 0.224411
## conditionmeaningfully 0.671 0.502715
## conditionmeaninglessly -0.809 0.419369
## conditionrepeat -1.027 0.305586
## beforeafteron 3.737 0.000232 ***
## agegroup2Y -0.719 0.472959
## conditioncontrol:beforeafteron -2.634 0.008992 **
## conditionmeaningfully:beforeafteron -1.614 0.107912
## conditionmeaninglessly:beforeafteron -2.325 0.020917 *
## conditionrepeat:beforeafteron -0.717 0.474250
## conditioncontrol:agegroup2Y 1.292 0.197584
## conditionmeaningfully:agegroup2Y -0.261 0.794688
## conditionmeaninglessly:agegroup2Y 0.316 0.751990
## conditionrepeat:agegroup2Y 0.976 0.329970
## beforeafteron:agegroup2Y -0.088 0.929644
## conditioncontrol:beforeafteron:agegroup2Y -0.273 0.784849
## conditionmeaningfully:beforeafteron:agegroup2Y 0.255 0.798681
## conditionmeaninglessly:beforeafteron:agegroup2Y 0.130 0.896803
## conditionrepeat:beforeafteron:agegroup2Y 0.481 0.631276
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Correlation matrix not shown by default, as p = 20 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 12.9560 3.2390 4 243 9.0838 7.453e-07
## beforeafter 11.0528 11.0528 1 243 30.9976 6.822e-08
## agegroup2 0.0024 0.0024 1 27 0.0066 0.9356637
## condition:beforeafter 8.4337 2.1084 4 243 5.9131 0.0001479
## condition:agegroup2 2.2923 0.5731 4 243 1.6072 0.1730818
## beforeafter:agegroup2 0.0112 0.0112 1 243 0.0313 0.8596271
## condition:beforeafter:agegroup2 0.2264 0.0566 4 243 0.1587 0.9588952
##
## condition ***
## beforeafter ***
## agegroup2
## condition:beforeafter ***
## condition:agegroup2
## beforeafter:agegroup2
## condition:beforeafter:agegroup2
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 SCL answer contr… 58 58 4.66 57 1.98e-5 1.98e-4 ***
## 2 SCL answer meani… 58 58 0.957 57 3.43e-1 1 e+0 ns
## 3 SCL answer meani… 58 58 4.34 57 5.80e-5 5.8 e-4 ***
## 4 SCL answer repeat 58 58 1.30 57 1.98e-1 1 e+0 ns
## 5 SCL control meani… 58 58 -3.28 57 2 e-3 1.8 e-2 *
## 6 SCL control meani… 58 58 -0.406 57 6.86e-1 1 e+0 ns
## 7 SCL control repeat 58 58 -3.55 57 7.72e-4 8 e-3 **
## 8 SCL meanin… meani… 58 58 2.69 57 9 e-3 9.2 e-2 ns
## 9 SCL meanin… repeat 58 58 0.0873 57 9.31e-1 1 e+0 ns
## 10 SCL meanin… repeat 58 58 -3.19 57 2 e-3 2.3 e-2 *## # A tibble: 1 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 SCL befor… on 145 145 -6.91 144 1.41e-10 1.41e-10 ****## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: SCL ~ condition + (1 | sub)
## Data: filter1.before
##
## REML criterion at convergence: 203.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.1254 -0.4916 -0.0964 0.2438 5.7312
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.006914 0.08315
## Residual 0.216155 0.46492
## Number of obs: 145, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 1.16091 0.08770 139.46402 13.237 <2e-16 ***
## conditioncontrol -0.06518 0.12210 112.00000 -0.534 0.595
## conditionmeaningfully 0.10921 0.12210 112.00000 0.894 0.373
## conditionmeaninglessly -0.13121 0.12210 112.00000 -1.075 0.285
## conditionrepeat -0.07328 0.12210 112.00000 -0.600 0.550
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.696
## cndtnmnngfl -0.696 0.500
## cndtnmnngls -0.696 0.500 0.500
## conditinrpt -0.696 0.500 0.500 0.500## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 0.97469 0.24367 4 112 1.1273 0.3473## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 SCL answer contr… 29 29 0.588 28 0.561 1 ns
## 2 SCL answer meani… 29 29 -0.864 28 0.395 1 ns
## 3 SCL answer meani… 29 29 1.32 28 0.197 1 ns
## 4 SCL answer repeat 29 29 0.786 28 0.438 1 ns
## 5 SCL control meani… 29 29 -1.17 28 0.252 1 ns
## 6 SCL control meani… 29 29 0.607 28 0.549 1 ns
## 7 SCL control repeat 29 29 0.0889 28 0.93 1 ns
## 8 SCL meaningful… meani… 29 29 1.48 28 0.151 1 ns
## 9 SCL meaningful… repeat 29 29 1.21 28 0.238 1 ns
## 10 SCL meaningles… repeat 29 29 -0.565 28 0.577 1 ns## ANOVA Table (type III tests)
##
## $ANOVA
## Effect DFn DFd F p p<.05 ges
## 1 condition 4 112 9.79 7.92e-07 * 0.212
##
## $`Mauchly's Test for Sphericity`
## Effect W p p<.05
## 1 condition 0.292 0.000168 *
##
## $`Sphericity Corrections`
## Effect GGe DF[GG] p[GG] p[GG]<.05 HFe DF[HF] p[HF]
## 1 condition 0.679 2.72, 76.07 2.93e-05 * 0.759 3.04, 85.03 1.19e-05
## p[HF]<.05
## 1 *## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: SCL ~ condition + (1 | sub)
## Data: filter1.on
##
## REML criterion at convergence: 327.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.7642 -0.6484 -0.0919 0.3802 4.3435
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.02093 0.1447
## Residual 0.51938 0.7207
## Number of obs: 145, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 1.9901 0.1365 139.1646 14.580 < 2e-16 ***
## conditioncontrol -0.9685 0.1893 112.0000 -5.117 1.30e-06 ***
## conditionmeaningfully -0.3472 0.1893 112.0000 -1.835 0.0692 .
## conditionmeaninglessly -0.8434 0.1893 112.0000 -4.456 1.99e-05 ***
## conditionrepeat -0.1917 0.1893 112.0000 -1.013 0.3132
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.693
## cndtnmnngfl -0.693 0.500
## cndtnmnngls -0.693 0.500 0.500
## conditinrpt -0.693 0.500 0.500 0.500## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 20.338 5.0846 4 112 9.7897 7.916e-07 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 SCL answer contr… 29 29 6.34 28 7.34e-7 7.34e-6 ****
## 2 SCL answer meani… 29 29 1.67 28 1.06e-1 1 e+0 ns
## 3 SCL answer meani… 29 29 4.69 28 6.41e-5 6.41e-4 ***
## 4 SCL answer repeat 29 29 1.05 28 3.02e-1 1 e+0 ns
## 5 SCL control meani… 29 29 -3.36 28 2 e-3 2.3 e-2 *
## 6 SCL control meani… 29 29 -1.31 28 2 e-1 1 e+0 ns
## 7 SCL control repeat 29 29 -4.62 28 7.93e-5 7.93e-4 ***
## 8 SCL meanin… meani… 29 29 2.25 28 3.2 e-2 3.21e-1 ns
## 9 SCL meanin… repeat 29 29 -0.578 28 5.68e-1 1 e+0 ns
## 10 SCL meanin… repeat 29 29 -3.56 28 1 e-3 1.4 e-2 *3.3.2 SDNN (Standard Deviation of Normal-to-Normal intervals (SDNN) extracted from photoplethysmography (PPG))
##
## Call:
## lm(formula = SDNN ~ condition * beforeafter, data = SDNNdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.11223 -0.28458 -0.07324 0.20317 2.08571
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.98759 0.08945 11.041 < 2e-16 ***
## conditioncontrol 0.22687 0.12649 1.794 0.073970 .
## conditionmeaningfully 0.15543 0.12649 1.229 0.220201
## conditionmeaninglessly 0.24946 0.12649 1.972 0.049581 *
## conditionrepeat 0.03465 0.12649 0.274 0.784363
## beforeafteron 0.49388 0.12649 3.904 0.000118 ***
## conditioncontrol:beforeafteron -0.69561 0.17889 -3.888 0.000126 ***
## conditionmeaningfully:beforeafteron -0.32284 0.17889 -1.805 0.072195 .
## conditionmeaninglessly:beforeafteron -0.38292 0.17889 -2.141 0.033174 *
## conditionrepeat:beforeafteron 0.13400 0.17889 0.749 0.454447
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.4817 on 280 degrees of freedom
## Multiple R-squared: 0.1568, Adjusted R-squared: 0.1297
## F-statistic: 5.784 on 9 and 280 DF, p-value: 2.213e-07
##
## Asymptotic one-sample Kolmogorov-Smirnov test
##
## data: res_lmmodel
## D = 0.10486, p-value = 0.003398
## alternative hypothesis: two-sided## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 9 3.1087 0.001401 **
## 280
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Call:
## lm(formula = log(SDNN) ~ condition * beforeafter, data = SDNNdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.28527 -0.21733 -0.00321 0.21114 0.98281
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.04754 0.06560 -0.725 0.469231
## conditioncontrol 0.16078 0.09277 1.733 0.084201 .
## conditionmeaningfully 0.13091 0.09277 1.411 0.159351
## conditionmeaninglessly 0.22165 0.09277 2.389 0.017552 *
## conditionrepeat 0.01976 0.09277 0.213 0.831491
## beforeafteron 0.33651 0.09277 3.627 0.000340 ***
## conditioncontrol:beforeafteron -0.48478 0.13120 -3.695 0.000265 ***
## conditionmeaningfully:beforeafteron -0.20253 0.13120 -1.544 0.123808
## conditionmeaninglessly:beforeafteron -0.27021 0.13120 -2.060 0.040369 *
## conditionrepeat:beforeafteron 0.10981 0.13120 0.837 0.403332
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3533 on 280 degrees of freedom
## Multiple R-squared: 0.1513, Adjusted R-squared: 0.124
## F-statistic: 5.546 on 9 and 280 DF, p-value: 4.865e-07
##
## Asymptotic one-sample Kolmogorov-Smirnov test
##
## data: res_lmmodel
## D = 0.041983, p-value = 0.6862
## alternative hypothesis: two-sided## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 9 1.4177 0.18
## 280## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: SDNN ~ condition * beforeafter * agegroup2 + (1 | sub)
## Data: SDNNdata
##
## REML criterion at convergence: 258.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.5096 -0.6263 -0.0239 0.5951 2.8714
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.01042 0.1021
## Residual 0.11739 0.3426
## Number of obs: 290, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df
## (Intercept) -0.07576 0.09555 254.76274
## conditioncontrol 0.13520 0.12950 243.00000
## conditionmeaningfully 0.08089 0.12950 243.00000
## conditionmeaninglessly 0.24911 0.12950 243.00000
## conditionrepeat 0.11007 0.12950 243.00000
## beforeafteron 0.35618 0.12950 243.00000
## agegroup2Y 0.05456 0.13285 254.76274
## conditioncontrol:beforeafteron -0.42650 0.18314 243.00000
## conditionmeaningfully:beforeafteron -0.17411 0.18314 243.00000
## conditionmeaninglessly:beforeafteron -0.31872 0.18314 243.00000
## conditionrepeat:beforeafteron 0.02315 0.18314 243.00000
## conditioncontrol:agegroup2Y 0.04945 0.18006 243.00000
## conditionmeaningfully:agegroup2Y 0.09670 0.18006 243.00000
## conditionmeaninglessly:agegroup2Y -0.05310 0.18006 243.00000
## conditionrepeat:agegroup2Y -0.17459 0.18006 243.00000
## beforeafteron:agegroup2Y -0.03802 0.18006 243.00000
## conditioncontrol:beforeafteron:agegroup2Y -0.11268 0.25464 243.00000
## conditionmeaningfully:beforeafteron:agegroup2Y -0.05494 0.25464 243.00000
## conditionmeaninglessly:beforeafteron:agegroup2Y 0.09379 0.25464 243.00000
## conditionrepeat:beforeafteron:agegroup2Y 0.16755 0.25464 243.00000
## t value Pr(>|t|)
## (Intercept) -0.793 0.4285
## conditioncontrol 1.044 0.2975
## conditionmeaningfully 0.625 0.5328
## conditionmeaninglessly 1.924 0.0556 .
## conditionrepeat 0.850 0.3962
## beforeafteron 2.750 0.0064 **
## agegroup2Y 0.411 0.6816
## conditioncontrol:beforeafteron -2.329 0.0207 *
## conditionmeaningfully:beforeafteron -0.951 0.3427
## conditionmeaninglessly:beforeafteron -1.740 0.0831 .
## conditionrepeat:beforeafteron 0.126 0.8995
## conditioncontrol:agegroup2Y 0.275 0.7838
## conditionmeaningfully:agegroup2Y 0.537 0.5917
## conditionmeaninglessly:agegroup2Y -0.295 0.7683
## conditionrepeat:agegroup2Y -0.970 0.3332
## beforeafteron:agegroup2Y -0.211 0.8329
## conditioncontrol:beforeafteron:agegroup2Y -0.443 0.6585
## conditionmeaningfully:beforeafteron:agegroup2Y -0.216 0.8293
## conditionmeaninglessly:beforeafteron:agegroup2Y 0.368 0.7130
## conditionrepeat:beforeafteron:agegroup2Y 0.658 0.5112
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Correlation matrix not shown by default, as p = 20 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 1.06345 0.26586 4 243 2.2648 0.06285
## beforeafter 2.02680 2.02680 1 243 17.2657 4.505e-05
## agegroup2 0.03140 0.03140 1 27 0.2675 0.60924
## condition:beforeafter 3.11118 0.77779 4 243 6.6258 4.466e-05
## condition:agegroup2 0.18661 0.04665 4 243 0.3974 0.81041
## beforeafter:agegroup2 0.00673 0.00673 1 243 0.0573 0.81095
## condition:beforeafter:agegroup2 0.18403 0.04601 4 243 0.3919 0.81433
##
## condition .
## beforeafter ***
## agegroup2
## condition:beforeafter ***
## condition:agegroup2
## beforeafter:agegroup2
## condition:beforeafter:agegroup2
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 1.06345 0.26586 4 243 2.2648 0.06285
## beforeafter 2.02680 2.02680 1 243 17.2657 4.505e-05
## agegroup2 0.03140 0.03140 1 27 0.2675 0.60924
## condition:beforeafter 3.11118 0.77779 4 243 6.6258 4.466e-05
## condition:agegroup2 0.18661 0.04665 4 243 0.3974 0.81041
## beforeafter:agegroup2 0.00673 0.00673 1 243 0.0573 0.81095
## condition:beforeafter:agegroup2 0.18403 0.04601 4 243 0.3919 0.81433
##
## condition .
## beforeafter ***
## agegroup2
## condition:beforeafter ***
## condition:agegroup2
## beforeafter:agegroup2
## condition:beforeafter:agegroup2
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## # A tibble: 1 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 SDNN beforei… on 145 145 -4.95 144 2.02e-6 2.02e-6 ****## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: SDNN ~ condition + (1 | sub)
## Data: filter1.before
##
## REML criterion at convergence: 96.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.85755 -0.57963 -0.01298 0.61145 2.91512
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.001433 0.03785
## Residual 0.102264 0.31979
## Number of obs: 145, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.04754 0.05980 139.89315 -0.795 0.42793
## conditioncontrol 0.16078 0.08398 112.00000 1.914 0.05811 .
## conditionmeaningfully 0.13091 0.08398 112.00000 1.559 0.12187
## conditionmeaninglessly 0.22165 0.08398 112.00000 2.639 0.00949 **
## conditionrepeat 0.01976 0.08398 112.00000 0.235 0.81441
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.702
## cndtnmnngfl -0.702 0.500
## cndtnmnngls -0.702 0.500 0.500
## conditinrpt -0.702 0.500 0.500 0.500## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 1.0343 0.25858 4 112 2.5286 0.04451 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 SDNN answer contr… 29 29 -1.88 28 0.07 0.701 ns
## 2 SDNN answer meani… 29 29 -1.73 28 0.095 0.954 ns
## 3 SDNN answer meani… 29 29 -2.64 28 0.013 0.133 ns
## 4 SDNN answer repeat 29 29 -0.260 28 0.797 1 ns
## 5 SDNN control meani… 29 29 0.337 28 0.739 1 ns
## 6 SDNN control meani… 29 29 -0.667 28 0.51 1 ns
## 7 SDNN control repeat 29 29 1.51 28 0.142 1 ns
## 8 SDNN meaningful… meani… 29 29 -1.15 28 0.26 1 ns
## 9 SDNN meaningful… repeat 29 29 1.38 28 0.178 1 ns
## 10 SDNN meaningles… repeat 29 29 2.40 28 0.023 0.234 ns## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: SDNN ~ condition + (1 | sub)
## Data: filter1.on
##
## REML criterion at convergence: 144.2
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.2811 -0.6132 -0.0445 0.5824 2.6031
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.005948 0.07713
## Residual 0.139960 0.37411
## Number of obs: 145, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.28897 0.07093 139.07542 4.074 7.73e-05 ***
## conditioncontrol -0.32400 0.09825 112.00000 -3.298 0.00131 **
## conditionmeaningfully -0.07162 0.09825 112.00000 -0.729 0.46752
## conditionmeaninglessly -0.04857 0.09825 112.00000 -0.494 0.62204
## conditionrepeat 0.12957 0.09825 112.00000 1.319 0.18992
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.693
## cndtnmnngfl -0.693 0.500
## cndtnmnngls -0.693 0.500 0.500
## conditinrpt -0.693 0.500 0.500 0.500## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 3.1743 0.79357 4 112 5.67 0.0003416 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 SDNN answer contr… 29 29 2.92 28 7 e-3 0.069 ns
## 2 SDNN answer meani… 29 29 0.816 28 4.21e-1 1 ns
## 3 SDNN answer meani… 29 29 0.438 28 6.65e-1 1 ns
## 4 SDNN answer repeat 29 29 -1.29 28 2.07e-1 1 ns
## 5 SDNN control meani… 29 29 -3.16 28 4 e-3 0.037 *
## 6 SDNN control meani… 29 29 -3.10 28 4 e-3 0.044 *
## 7 SDNN control repeat 29 29 -3.86 28 6.04e-4 0.006 **
## 8 SDNN meaningf… meani… 29 29 -0.283 28 7.79e-1 1 ns
## 9 SDNN meaningf… repeat 29 29 -2.11 28 4.4 e-2 0.44 ns
## 10 SDNN meaningl… repeat 29 29 -1.75 28 9.1 e-2 0.909 ns3.3.3 Breathing rate
##
## Call:
## lm(formula = RESP ~ condition * beforeafter, data = RESPdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.93300 -0.13385 -0.00261 0.11281 1.42227
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.91290 0.04784 19.081 < 2e-16 ***
## conditioncontrol 0.11058 0.06766 1.634 0.10331
## conditionmeaningfully 0.11154 0.06766 1.649 0.10037
## conditionmeaninglessly 0.10177 0.06766 1.504 0.13368
## conditionrepeat 0.18376 0.06766 2.716 0.00702 **
## beforeafteron 0.04912 0.06766 0.726 0.46851
## conditioncontrol:beforeafteron -0.06190 0.09569 -0.647 0.51820
## conditionmeaningfully:beforeafteron -0.12551 0.09569 -1.312 0.19071
## conditionmeaninglessly:beforeafteron -0.11834 0.09569 -1.237 0.21722
## conditionrepeat:beforeafteron -0.04082 0.09569 -0.427 0.67004
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2577 on 280 degrees of freedom
## Multiple R-squared: 0.05368, Adjusted R-squared: 0.02326
## F-statistic: 1.765 on 9 and 280 DF, p-value: 0.07481
##
## Asymptotic one-sample Kolmogorov-Smirnov test
##
## data: res_lmmodel
## D = 0.11272, p-value = 0.00126
## alternative hypothesis: two-sided## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 9 1.3775 0.1979
## 280##
## Call:
## lm(formula = log(RESP) ~ condition * beforeafter, data = RESPdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.79285 -0.11205 0.02751 0.13633 0.89478
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.12078 0.05298 -2.280 0.0234 *
## conditioncontrol 0.12720 0.07492 1.698 0.0907 .
## conditionmeaningfully 0.12467 0.07492 1.664 0.0972 .
## conditionmeaninglessly 0.10573 0.07492 1.411 0.1593
## conditionrepeat 0.15969 0.07492 2.131 0.0339 *
## beforeafteron 0.04555 0.07492 0.608 0.5437
## conditioncontrol:beforeafteron -0.06159 0.10595 -0.581 0.5615
## conditionmeaningfully:beforeafteron -0.14861 0.10595 -1.403 0.1619
## conditionmeaninglessly:beforeafteron -0.10577 0.10595 -0.998 0.3190
## conditionrepeat:beforeafteron -0.05212 0.10595 -0.492 0.6231
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2853 on 280 degrees of freedom
## Multiple R-squared: 0.03541, Adjusted R-squared: 0.004407
## F-statistic: 1.142 on 9 and 280 DF, p-value: 0.3329
##
## Asymptotic one-sample Kolmogorov-Smirnov test
##
## data: res_lmmodel
## D = 0.124, p-value = 0.0002678
## alternative hypothesis: two-sided## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 9 0.7143 0.6957
## 280## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RESP ~ condition * beforeafter * agegroup2 + (1 | sub)
## Data: RESPdata
##
## REML criterion at convergence: 126.9
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -5.7583 -0.3806 0.0754 0.5531 3.3718
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.01330 0.1153
## Residual 0.06901 0.2627
## Number of obs: 290, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error
## (Intercept) -0.122394 0.076677
## conditioncontrol 0.125243 0.099292
## conditionmeaningfully 0.121261 0.099292
## conditionmeaninglessly 0.054179 0.099292
## conditionrepeat 0.130165 0.099292
## beforeafteron 0.105056 0.099292
## agegroup2Y 0.003128 0.106615
## conditioncontrol:beforeafteron -0.146273 0.140420
## conditionmeaningfully:beforeafteron -0.120484 0.140420
## conditionmeaninglessly:beforeafteron -0.110818 0.140420
## conditionrepeat:beforeafteron -0.158310 0.140420
## conditioncontrol:agegroup2Y 0.003784 0.138060
## conditionmeaningfully:agegroup2Y 0.006593 0.138060
## conditionmeaninglessly:agegroup2Y 0.099667 0.138060
## conditionrepeat:agegroup2Y 0.057085 0.138060
## beforeafteron:agegroup2Y -0.115041 0.138060
## conditioncontrol:beforeafteron:agegroup2Y 0.163710 0.195246
## conditionmeaningfully:beforeafteron:agegroup2Y -0.054370 0.195246
## conditionmeaninglessly:beforeafteron:agegroup2Y 0.009764 0.195246
## conditionrepeat:beforeafteron:agegroup2Y 0.205291 0.195246
## df t value Pr(>|t|)
## (Intercept) 218.635364 -1.596 0.112
## conditioncontrol 243.000001 1.261 0.208
## conditionmeaningfully 243.000001 1.221 0.223
## conditionmeaninglessly 243.000001 0.546 0.586
## conditionrepeat 243.000001 1.311 0.191
## beforeafteron 243.000001 1.058 0.291
## agegroup2Y 218.635364 0.029 0.977
## conditioncontrol:beforeafteron 243.000001 -1.042 0.299
## conditionmeaningfully:beforeafteron 243.000001 -0.858 0.392
## conditionmeaninglessly:beforeafteron 243.000001 -0.789 0.431
## conditionrepeat:beforeafteron 243.000001 -1.127 0.261
## conditioncontrol:agegroup2Y 243.000001 0.027 0.978
## conditionmeaningfully:agegroup2Y 243.000001 0.048 0.962
## conditionmeaninglessly:agegroup2Y 243.000001 0.722 0.471
## conditionrepeat:agegroup2Y 243.000001 0.413 0.680
## beforeafteron:agegroup2Y 243.000001 -0.833 0.406
## conditioncontrol:beforeafteron:agegroup2Y 243.000001 0.838 0.403
## conditionmeaningfully:beforeafteron:agegroup2Y 243.000001 -0.278 0.781
## conditionmeaninglessly:beforeafteron:agegroup2Y 243.000001 0.050 0.960
## conditionrepeat:beforeafteron:agegroup2Y 243.000001 1.051 0.294##
## Correlation matrix not shown by default, as p = 20 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 0.57105 0.142762 4 243 2.0686 0.08554 .
## beforeafter 0.05358 0.053581 1 243 0.7764 0.37911
## agegroup2 0.00326 0.003256 1 27 0.0472 0.82967
## condition:beforeafter 0.17884 0.044710 4 243 0.6478 0.62891
## condition:agegroup2 0.32602 0.081506 4 243 1.1810 0.31971
## beforeafter:agegroup2 0.04555 0.045552 1 243 0.6600 0.41734
## condition:beforeafter:agegroup2 0.18448 0.046119 4 243 0.6683 0.61458
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 13.4 Eye movement indicators
3.4.1 Pupil diameter (PD)
##
## Call:
## lm(formula = PD ~ condition * beforeafter, data = PDdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.167581 -0.049885 0.000388 0.041872 0.304631
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.999013 0.013789 72.451 < 2e-16 ***
## conditioncontrol 0.002304 0.019500 0.118 0.906
## conditionmeaningfully -0.001545 0.019500 -0.079 0.937
## conditionmeaninglessly 0.011311 0.019500 0.580 0.562
## conditionrepeat -0.008317 0.019500 -0.427 0.670
## beforeafteron 0.125300 0.019500 6.426 5.63e-10 ***
## conditioncontrol:beforeafteron -0.129992 0.027578 -4.714 3.84e-06 ***
## conditionmeaningfully:beforeafteron -0.113723 0.027578 -4.124 4.92e-05 ***
## conditionmeaninglessly:beforeafteron -0.136238 0.027578 -4.940 1.34e-06 ***
## conditionrepeat:beforeafteron -0.037815 0.027578 -1.371 0.171
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.07426 on 280 degrees of freedom
## Multiple R-squared: 0.2485, Adjusted R-squared: 0.2243
## F-statistic: 10.29 on 9 and 280 DF, p-value: 1.042e-13
##
## Asymptotic one-sample Kolmogorov-Smirnov test
##
## data: res_lmmodel
## D = 0.036975, p-value = 0.8227
## alternative hypothesis: two-sided## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 9 0.6456 0.7576
## 280## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: PD ~ agegroup2 * condition * beforeafter + (1 | sub)
## Data: PDdata
##
## REML criterion at convergence: -611
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.1952 -0.5610 0.0032 0.6031 3.9362
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.001258 0.03547
## Residual 0.004362 0.06605
## Number of obs: 290, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error
## (Intercept) 9.994e-01 2.004e-02
## agegroup2Y -6.955e-04 2.786e-02
## conditioncontrol -1.054e-02 2.496e-02
## conditionmeaningfully 1.781e-03 2.496e-02
## conditionmeaninglessly 1.882e-02 2.496e-02
## conditionrepeat -2.135e-02 2.496e-02
## beforeafteron 1.257e-01 2.496e-02
## agegroup2Y:conditioncontrol 2.484e-02 3.471e-02
## agegroup2Y:conditionmeaningfully -6.429e-03 3.471e-02
## agegroup2Y:conditionmeaninglessly -1.452e-02 3.471e-02
## agegroup2Y:conditionrepeat 2.519e-02 3.471e-02
## agegroup2Y:beforeafteron -8.193e-04 3.471e-02
## conditioncontrol:beforeafteron -1.167e-01 3.530e-02
## conditionmeaningfully:beforeafteron -1.233e-01 3.530e-02
## conditionmeaninglessly:beforeafteron -1.211e-01 3.530e-02
## conditionrepeat:beforeafteron -2.537e-02 3.530e-02
## agegroup2Y:conditioncontrol:beforeafteron -2.574e-02 4.909e-02
## agegroup2Y:conditionmeaningfully:beforeafteron 1.845e-02 4.909e-02
## agegroup2Y:conditionmeaninglessly:beforeafteron -2.927e-02 4.909e-02
## agegroup2Y:conditionrepeat:beforeafteron -2.406e-02 4.909e-02
## df t value Pr(>|t|)
## (Intercept) 1.861e+02 49.877 < 2e-16 ***
## agegroup2Y 1.861e+02 -0.025 0.980109
## conditioncontrol 2.430e+02 -0.422 0.673113
## conditionmeaningfully 2.430e+02 0.071 0.943186
## conditionmeaninglessly 2.430e+02 0.754 0.451637
## conditionrepeat 2.430e+02 -0.855 0.393335
## beforeafteron 2.430e+02 5.036 9.26e-07 ***
## agegroup2Y:conditioncontrol 2.430e+02 0.716 0.474900
## agegroup2Y:conditionmeaningfully 2.430e+02 -0.185 0.853204
## agegroup2Y:conditionmeaninglessly 2.430e+02 -0.418 0.676146
## agegroup2Y:conditionrepeat 2.430e+02 0.726 0.468706
## agegroup2Y:beforeafteron 2.430e+02 -0.024 0.981188
## conditioncontrol:beforeafteron 2.430e+02 -3.305 0.001093 **
## conditionmeaningfully:beforeafteron 2.430e+02 -3.492 0.000570 ***
## conditionmeaninglessly:beforeafteron 2.430e+02 -3.430 0.000709 ***
## conditionrepeat:beforeafteron 2.430e+02 -0.719 0.473034
## agegroup2Y:conditioncontrol:beforeafteron 2.430e+02 -0.524 0.600549
## agegroup2Y:conditionmeaningfully:beforeafteron 2.430e+02 0.376 0.707338
## agegroup2Y:conditionmeaninglessly:beforeafteron 2.430e+02 -0.596 0.551493
## agegroup2Y:conditionrepeat:beforeafteron 2.430e+02 -0.490 0.624518
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Correlation matrix not shown by default, as p = 20 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## agegroup2 0.000034 0.000034 1 27 0.0078 0.9303
## condition 0.167015 0.041754 4 243 9.5715 3.339e-07
## beforeafter 0.127553 0.127553 1 243 29.2399 1.528e-07
## agegroup2:condition 0.017003 0.004251 4 243 0.9744 0.4221
## agegroup2:beforeafter 0.003033 0.003033 1 243 0.6952 0.4052
## condition:beforeafter 0.215486 0.053872 4 243 12.3493 3.701e-09
## agegroup2:condition:beforeafter 0.006168 0.001542 4 243 0.3535 0.8415
##
## agegroup2
## condition ***
## beforeafter ***
## agegroup2:condition
## agegroup2:beforeafter
## condition:beforeafter ***
## agegroup2:condition:beforeafter
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 PD answer contr… 58 58 4.02 57 1.74e-4 0.002 **
## 2 PD answer meani… 58 58 3.79 57 3.67e-4 0.004 **
## 3 PD answer meani… 58 58 3.80 57 3.56e-4 0.004 **
## 4 PD answer repeat 58 58 2.04 57 4.6 e-2 0.457 ns
## 5 PD control meani… 58 58 -0.385 57 7.02e-1 1 ns
## 6 PD control meani… 58 58 -0.572 57 5.7 e-1 1 ns
## 7 PD control repeat 58 58 -2.67 57 1 e-2 0.1 ns
## 8 PD meaningf… meani… 58 58 -0.127 57 8.99e-1 1 ns
## 9 PD meaningf… repeat 58 58 -2.71 57 9 e-3 0.088 ns
## 10 PD meaningl… repeat 58 58 -2.05 57 4.5 e-2 0.447 ns## # A tibble: 1 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 PD beforei… on 145 145 -5.48 144 1.88e-7 1.88e-7 ****## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: PD ~ condition + (1 | sub)
## Data: filter1.before
##
## REML criterion at convergence: -352.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.98686 -0.57781 0.02829 0.59355 2.56295
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.0006426 0.02535
## Residual 0.0036845 0.06070
## Number of obs: 145, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.999013 0.012215 128.651048 81.784 <2e-16 ***
## conditioncontrol 0.002304 0.015941 112.000000 0.145 0.885
## conditionmeaningfully -0.001545 0.015941 112.000000 -0.097 0.923
## conditionmeaninglessly 0.011311 0.015941 112.000000 0.710 0.479
## conditionrepeat -0.008317 0.015941 112.000000 -0.522 0.603
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.652
## cndtnmnngfl -0.652 0.500
## cndtnmnngls -0.652 0.500 0.500
## conditinrpt -0.652 0.500 0.500 0.500## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 0.0058579 0.0014645 4 112 0.3975 0.8101## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 PD answer contr… 29 29 -0.134 28 0.894 1 ns
## 2 PD answer meani… 29 29 0.0927 28 0.927 1 ns
## 3 PD answer meani… 29 29 -0.723 28 0.476 1 ns
## 4 PD answer repeat 29 29 0.473 28 0.64 1 ns
## 5 PD control meani… 29 29 0.247 28 0.806 1 ns
## 6 PD control meani… 29 29 -0.597 28 0.555 1 ns
## 7 PD control repeat 29 29 0.715 28 0.481 1 ns
## 8 PD meaningful… meani… 29 29 -0.744 28 0.463 1 ns
## 9 PD meaningful… repeat 29 29 0.590 28 0.56 1 ns
## 10 PD meaningles… repeat 29 29 1.14 28 0.263 1 ns## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: PD ~ condition + (1 | sub)
## Data: filter1.on
##
## REML criterion at convergence: -300.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.0153 -0.5500 -0.0672 0.5965 3.4668
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.001772 0.0421
## Residual 0.004928 0.0702
## Number of obs: 145, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 1.12431 0.01520 109.39429 73.966 < 2e-16 ***
## conditioncontrol -0.12769 0.01844 112.00000 -6.926 2.88e-10 ***
## conditionmeaningfully -0.11527 0.01844 112.00000 -6.252 7.56e-09 ***
## conditionmeaninglessly -0.12493 0.01844 112.00000 -6.776 6.03e-10 ***
## conditionrepeat -0.04613 0.01844 112.00000 -2.502 0.0138 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.606
## cndtnmnngfl -0.606 0.500
## cndtnmnngls -0.606 0.500 0.500
## conditinrpt -0.606 0.500 0.500 0.500## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 0.37828 0.09457 4 112 19.188 4.81e-12 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 PD answer contr… 29 29 6.43 28 5.87e-7 5.87e-6 ****
## 2 PD answer meani… 29 29 5.38 28 9.84e-6 9.84e-5 ****
## 3 PD answer meani… 29 29 6.83 28 2.01e-7 2.01e-6 ****
## 4 PD answer repeat 29 29 2.34 28 2.7 e-2 2.67e-1 ns
## 5 PD control meani… 29 29 -0.774 28 4.45e-1 1 e+0 ns
## 6 PD control meani… 29 29 -0.194 28 8.48e-1 1 e+0 ns
## 7 PD control repeat 29 29 -4.37 28 1.54e-4 2 e-3 **
## 8 PD meanin… meani… 29 29 0.529 28 6.01e-1 1 e+0 ns
## 9 PD meanin… repeat 29 29 -3.97 28 4.54e-4 5 e-3 **
## 10 PD meanin… repeat 29 29 -4.06 28 3.62e-4 4 e-3 **3.4.2 PVRC (proportion of visit duration on the road centre)
##
## Call:
## lm(formula = PVRC ~ condition * beforeafter, data = PVRCdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.76132 -0.17676 0.00025 0.13603 1.66148
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.84328 0.06095 13.836 <2e-16 ***
## conditioncontrol 0.18364 0.08620 2.130 0.0340 *
## conditionmeaningfully 0.10292 0.08620 1.194 0.2335
## conditionmeaninglessly 0.06242 0.08620 0.724 0.4696
## conditionrepeat 0.04501 0.08620 0.522 0.6019
## beforeafteron 0.15329 0.08620 1.778 0.0764 .
## conditioncontrol:beforeafteron -0.30219 0.12190 -2.479 0.0138 *
## conditionmeaningfully:beforeafteron -0.09009 0.12190 -0.739 0.4605
## conditionmeaninglessly:beforeafteron -0.08303 0.12190 -0.681 0.4964
## conditionrepeat:beforeafteron 0.08218 0.12190 0.674 0.5008
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.3282 on 280 degrees of freedom
## Multiple R-squared: 0.05792, Adjusted R-squared: 0.02764
## F-statistic: 1.913 on 9 and 280 DF, p-value: 0.05009
## Warning in ks.test.default(res_lmmodel, "pnorm", mean(res_lmmodel),
## sd(res_lmmodel)): ties should not be present for the one-sample
## Kolmogorov-Smirnov test##
## Asymptotic one-sample Kolmogorov-Smirnov test
##
## data: res_lmmodel
## D = 0.093022, p-value = 0.01323
## alternative hypothesis: two-sided## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 9 0.3091 0.9716
## 280## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: PVRC ~ condition * beforeafter * agegroup2 + (1 | sub)
## Data: PVRCdata
##
## REML criterion at convergence: 216.1
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.5205 -0.4645 -0.0257 0.4468 4.7174
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.007596 0.08715
## Residual 0.101093 0.31795
## Number of obs: 290, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error
## (Intercept) 0.826389 0.088111
## conditioncontrol 0.113661 0.120175
## conditionmeaningfully 0.055743 0.120175
## conditionmeaninglessly 0.029065 0.120175
## conditionrepeat 0.009350 0.120175
## beforeafteron 0.150082 0.120175
## agegroup2Y 0.032664 0.122513
## conditioncontrol:beforeafteron -0.251911 0.169952
## conditionmeaningfully:beforeafteron -0.029725 0.169952
## conditionmeaninglessly:beforeafteron 0.024860 0.169952
## conditionrepeat:beforeafteron 0.048359 0.169952
## conditioncontrol:agegroup2Y 0.135291 0.167096
## conditionmeaningfully:agegroup2Y 0.091219 0.167096
## conditionmeaninglessly:agegroup2Y 0.064491 0.167096
## conditionrepeat:agegroup2Y 0.068948 0.167096
## beforeafteron:agegroup2Y 0.006205 0.167096
## conditioncontrol:beforeafteron:agegroup2Y -0.097203 0.236309
## conditionmeaningfully:beforeafteron:agegroup2Y -0.116706 0.236309
## conditionmeaninglessly:beforeafteron:agegroup2Y -0.208583 0.236309
## conditionrepeat:beforeafteron:agegroup2Y 0.065388 0.236309
## df t value Pr(>|t|)
## (Intercept) 258.632400 9.379 <2e-16 ***
## conditioncontrol 243.000000 0.946 0.345
## conditionmeaningfully 243.000000 0.464 0.643
## conditionmeaninglessly 243.000000 0.242 0.809
## conditionrepeat 243.000000 0.078 0.938
## beforeafteron 243.000000 1.249 0.213
## agegroup2Y 258.632400 0.267 0.790
## conditioncontrol:beforeafteron 243.000000 -1.482 0.140
## conditionmeaningfully:beforeafteron 243.000000 -0.175 0.861
## conditionmeaninglessly:beforeafteron 243.000000 0.146 0.884
## conditionrepeat:beforeafteron 243.000000 0.285 0.776
## conditioncontrol:agegroup2Y 243.000000 0.810 0.419
## conditionmeaningfully:agegroup2Y 243.000000 0.546 0.586
## conditionmeaninglessly:agegroup2Y 243.000000 0.386 0.700
## conditionrepeat:agegroup2Y 243.000000 0.413 0.680
## beforeafteron:agegroup2Y 243.000000 0.037 0.970
## conditioncontrol:beforeafteron:agegroup2Y 243.000000 -0.411 0.681
## conditionmeaningfully:beforeafteron:agegroup2Y 243.000000 -0.494 0.622
## conditionmeaninglessly:beforeafteron:agegroup2Y 243.000000 -0.883 0.378
## conditionrepeat:beforeafteron:agegroup2Y 243.000000 0.277 0.782
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Correlation matrix not shown by default, as p = 20 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 0.24789 0.06197 4 243 0.6130 0.6536
## beforeafter 0.41596 0.41596 1 243 4.1146 0.0436 *
## agegroup2 0.21462 0.21462 1 27 2.1230 0.1566
## condition:beforeafter 1.17304 0.29326 4 243 2.9009 0.0226 *
## condition:agegroup2 0.20174 0.05043 4 243 0.4989 0.7366
## beforeafter:agegroup2 0.07700 0.07700 1 243 0.7616 0.3837
## condition:beforeafter:agegroup2 0.16419 0.04105 4 243 0.4060 0.8042
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## # A tibble: 1 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 PVRC beforeinter… on 145 145 -2.86 144 0.005 0.005 **## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: PVRC ~ condition + (1 | sub)
## Data: filter1.before
##
## REML criterion at convergence: 93.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.4045 -0.5141 -0.0425 0.3944 3.7727
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.007584 0.08709
## Residual 0.094516 0.30743
## Number of obs: 145, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.84328 0.05934 136.97660 14.212 <2e-16 ***
## conditioncontrol 0.18364 0.08074 112.00000 2.275 0.0248 *
## conditionmeaningfully 0.10292 0.08074 112.00000 1.275 0.2050
## conditionmeaninglessly 0.06242 0.08074 112.00000 0.773 0.4411
## conditionrepeat 0.04501 0.08074 112.00000 0.558 0.5783
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.680
## cndtnmnngfl -0.680 0.500
## cndtnmnngls -0.680 0.500 0.500
## conditinrpt -0.680 0.500 0.500 0.500## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 0.55658 0.13914 4 112 1.4722 0.2154## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 PVRC answer contr… 29 29 -2.13 28 0.042 0.424 ns
## 2 PVRC answer meani… 29 29 -1.31 28 0.201 1 ns
## 3 PVRC answer meani… 29 29 -0.772 28 0.447 1 ns
## 4 PVRC answer repeat 29 29 -0.570 28 0.573 1 ns
## 5 PVRC control meani… 29 29 0.842 28 0.407 1 ns
## 6 PVRC control meani… 29 29 1.51 28 0.143 1 ns
## 7 PVRC control repeat 29 29 1.51 28 0.142 1 ns
## 8 PVRC meaningful… meani… 29 29 0.610 28 0.547 1 ns
## 9 PVRC meaningful… repeat 29 29 0.818 28 0.421 1 ns
## 10 PVRC meaningles… repeat 29 29 0.240 28 0.812 1 ns## boundary (singular) fit: see help('isSingular')## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: PVRC ~ condition + (1 | sub)
## Data: filter1.on
##
## REML criterion at convergence: 109.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.1262 -0.5291 0.0013 0.4067 4.9347
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 1.575e-20 1.255e-10
## Residual 1.134e-01 3.367e-01
## Number of obs: 145, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.99658 0.06252 140.00000 15.939 <2e-16 ***
## conditioncontrol -0.11855 0.08842 140.00000 -1.341 0.182
## conditionmeaningfully 0.01283 0.08842 140.00000 0.145 0.885
## conditionmeaninglessly -0.02061 0.08842 140.00000 -0.233 0.816
## conditionrepeat 0.12719 0.08842 140.00000 1.439 0.153
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.707
## cndtnmnngfl -0.707 0.500
## cndtnmnngls -0.707 0.500 0.500
## conditinrpt -0.707 0.500 0.500 0.500
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 0.89381 0.22345 4 140 1.9711 0.1022## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 PVRC answer contr… 29 29 1.26 28 0.218 1 ns
## 2 PVRC answer meani… 29 29 -0.190 28 0.851 1 ns
## 3 PVRC answer meani… 29 29 0.249 28 0.806 1 ns
## 4 PVRC answer repeat 29 29 -1.46 28 0.155 1 ns
## 5 PVRC control meani… 29 29 -1.25 28 0.221 1 ns
## 6 PVRC control meani… 29 29 -1.14 28 0.265 1 ns
## 7 PVRC control repeat 29 29 -2.29 28 0.03 0.297 ns
## 8 PVRC meaningful… meani… 29 29 0.434 28 0.668 1 ns
## 9 PVRC meaningful… repeat 29 29 -1.32 28 0.196 1 ns
## 10 PVRC meaningles… repeat 29 29 -1.67 28 0.106 1 ns3.4.3 Blink rate (BL)
##
## Call:
## lm(formula = BL ~ condition * beforeafter, data = BLdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -1.6117 -0.5734 -0.2577 0.1255 6.4512
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.36109 0.19852 6.856 4.51e-11 ***
## conditioncontrol -0.07030 0.28075 -0.250 0.802
## conditionmeaningfully 0.28521 0.28075 1.016 0.311
## conditionmeaninglessly -0.01145 0.28075 -0.041 0.968
## conditionrepeat 0.29991 0.28075 1.068 0.286
## beforeafteron 0.28710 0.28075 1.023 0.307
## conditioncontrol:beforeafteron -0.33650 0.39704 -0.848 0.397
## conditionmeaningfully:beforeafteron -0.26692 0.39704 -0.672 0.502
## conditionmeaninglessly:beforeafteron -0.26211 0.39704 -0.660 0.510
## conditionrepeat:beforeafteron -0.26186 0.39704 -0.660 0.510
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.069 on 280 degrees of freedom
## Multiple R-squared: 0.02642, Adjusted R-squared: -0.004878
## F-statistic: 0.8441 on 9 and 280 DF, p-value: 0.576
## Warning in ks.test.default(res_lmmodel, "pnorm", mean(res_lmmodel),
## sd(res_lmmodel)): ties should not be present for the one-sample
## Kolmogorov-Smirnov test##
## Asymptotic one-sample Kolmogorov-Smirnov test
##
## data: res_lmmodel
## D = 0.20453, p-value = 5.806e-11
## alternative hypothesis: two-sided## Levene's Test for Homogeneity of Variance (center = median)
## Df F value Pr(>F)
## group 9 0.9794 0.4572
## 280##
## Call:
## lm(formula = log(BL) ~ condition * beforeafter, data = BLdata)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.1785 -0.3080 -0.0386 0.2462 1.9252
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.198710 0.106273 1.870 0.0626 .
## conditioncontrol -0.086689 0.150293 -0.577 0.5645
## conditionmeaningfully 0.139745 0.150293 0.930 0.3533
## conditionmeaninglessly -0.006716 0.150293 -0.045 0.9644
## conditionrepeat -0.030495 0.150293 -0.203 0.8394
## beforeafteron 0.130059 0.150293 0.865 0.3876
## conditioncontrol:beforeafteron -0.131528 0.212546 -0.619 0.5365
## conditionmeaningfully:beforeafteron -0.109103 0.212546 -0.513 0.6081
## conditionmeaninglessly:beforeafteron -0.135764 0.212546 -0.639 0.5235
## conditionrepeat:beforeafteron -0.007019 0.212546 -0.033 0.9737
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.5723 on 280 degrees of freedom
## Multiple R-squared: 0.02417, Adjusted R-squared: -0.007193
## F-statistic: 0.7707 on 9 and 280 DF, p-value: 0.6437
## Warning in ks.test.default(res_lmmodel, "pnorm", mean(res_lmmodel),
## sd(res_lmmodel)): ties should not be present for the one-sample
## Kolmogorov-Smirnov test##
## Asymptotic one-sample Kolmogorov-Smirnov test
##
## data: res_lmmodel
## D = 0.087813, p-value = 0.02284
## alternative hypothesis: two-sided## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: BL ~ condition * beforeafter * agegroup2 + (1 | sub)
## Data: BLdata
##
## REML criterion at convergence: 499
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -5.5925 -0.5150 -0.0064 0.4319 3.1751
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.0622 0.2494
## Residual 0.2706 0.5202
## Number of obs: 290, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error
## (Intercept) 0.179747 0.154189
## conditioncontrol -0.084538 0.196628
## conditionmeaningfully 0.191232 0.196628
## conditionmeaninglessly 0.127925 0.196628
## conditionrepeat 0.093849 0.196628
## beforeafteron 0.179699 0.196628
## agegroup2Y 0.036662 0.214391
## conditioncontrol:beforeafteron -0.197396 0.278074
## conditionmeaningfully:beforeafteron -0.205694 0.278074
## conditionmeaninglessly:beforeafteron -0.297859 0.278074
## conditionrepeat:beforeafteron 0.036842 0.278074
## conditioncontrol:agegroup2Y -0.004159 0.273400
## conditionmeaningfully:agegroup2Y -0.099541 0.273400
## conditionmeaninglessly:agegroup2Y -0.260306 0.273400
## conditionrepeat:agegroup2Y -0.240399 0.273400
## beforeafteron:agegroup2Y -0.095971 0.273400
## conditioncontrol:beforeafteron:agegroup2Y 0.127345 0.386646
## conditionmeaningfully:beforeafteron:agegroup2Y 0.186744 0.386646
## conditionmeaninglessly:beforeafteron:agegroup2Y 0.313383 0.386646
## conditionrepeat:beforeafteron:agegroup2Y -0.084798 0.386646
## df t value Pr(>|t|)
## (Intercept) 205.430939 1.166 0.245
## conditioncontrol 243.000000 -0.430 0.668
## conditionmeaningfully 243.000000 0.973 0.332
## conditionmeaninglessly 243.000000 0.651 0.516
## conditionrepeat 243.000000 0.477 0.634
## beforeafteron 243.000000 0.914 0.362
## agegroup2Y 205.430939 0.171 0.864
## conditioncontrol:beforeafteron 243.000000 -0.710 0.478
## conditionmeaningfully:beforeafteron 243.000000 -0.740 0.460
## conditionmeaninglessly:beforeafteron 243.000000 -1.071 0.285
## conditionrepeat:beforeafteron 243.000000 0.132 0.895
## conditioncontrol:agegroup2Y 243.000000 -0.015 0.988
## conditionmeaningfully:agegroup2Y 243.000000 -0.364 0.716
## conditionmeaninglessly:agegroup2Y 243.000000 -0.952 0.342
## conditionrepeat:agegroup2Y 243.000000 -0.879 0.380
## beforeafteron:agegroup2Y 243.000000 -0.351 0.726
## conditioncontrol:beforeafteron:agegroup2Y 243.000000 0.329 0.742
## conditionmeaningfully:beforeafteron:agegroup2Y 243.000000 0.483 0.630
## conditionmeaninglessly:beforeafteron:agegroup2Y 243.000000 0.811 0.418
## conditionrepeat:beforeafteron:agegroup2Y 243.000000 -0.219 0.827##
## Correlation matrix not shown by default, as p = 20 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 1.80709 0.45177 4 243 1.6693 0.1577
## beforeafter 0.20464 0.20464 1 243 0.7561 0.3854
## agegroup2 0.13336 0.13336 1 27 0.4928 0.4887
## condition:beforeafter 0.28391 0.07098 4 243 0.2623 0.9020
## condition:agegroup2 1.04426 0.26107 4 243 0.9646 0.4275
## beforeafter:agegroup2 0.00286 0.00286 1 243 0.0106 0.9182
## condition:beforeafter:agegroup2 0.35335 0.08834 4 243 0.3264 0.8601## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 1.80709 0.45177 4 243 1.6693 0.1577
## beforeafter 0.20464 0.20464 1 243 0.7561 0.3854
## agegroup2 0.13336 0.13336 1 27 0.4928 0.4887
## condition:beforeafter 0.28391 0.07098 4 243 0.2623 0.9020
## condition:agegroup2 1.04426 0.26107 4 243 0.9646 0.4275
## beforeafter:agegroup2 0.00286 0.00286 1 243 0.0106 0.9182
## condition:beforeafter:agegroup2 0.35335 0.08834 4 243 0.3264 0.8601## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 BL answer contr… 58 58 1.87 57 0.067 0.667 ns
## 2 BL answer meani… 58 58 -0.875 57 0.385 1 ns
## 3 BL answer meani… 58 58 0.854 57 0.397 1 ns
## 4 BL answer repeat 58 58 0.347 57 0.73 1 ns
## 5 BL control meani… 58 58 -3.03 57 0.004 0.037 *
## 6 BL control meani… 58 58 -1.17 57 0.245 1 ns
## 7 BL control repeat 58 58 -1.02 57 0.31 1 ns
## 8 BL meaningful… meani… 58 58 1.76 57 0.084 0.844 ns
## 9 BL meaningful… repeat 58 58 1.05 57 0.299 1 ns
## 10 BL meaningles… repeat 58 58 -0.338 57 0.737 1 ns## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: BL ~ condition + (1 | sub)
## Data: filter1.before
##
## REML criterion at convergence: 261.7
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -5.1225 -0.5119 -0.0390 0.4054 3.3590
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.04993 0.2235
## Residual 0.29797 0.5459
## Number of obs: 145, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.198710 0.109528 129.343102 1.814 0.072 .
## conditioncontrol -0.086689 0.143351 112.000000 -0.605 0.547
## conditionmeaningfully 0.139745 0.143351 112.000000 0.975 0.332
## conditionmeaninglessly -0.006716 0.143351 112.000000 -0.047 0.963
## conditionrepeat -0.030495 0.143351 112.000000 -0.213 0.832
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.654
## cndtnmnngfl -0.654 0.500
## cndtnmnngls -0.654 0.500 0.500
## conditinrpt -0.654 0.500 0.500 0.500## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 0.81109 0.20277 4 112 0.6805 0.6069## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 BL answer contr… 29 29 0.735 28 0.468 1 ns
## 2 BL answer meani… 29 29 -1.09 28 0.285 1 ns
## 3 BL answer meani… 29 29 0.0628 28 0.95 1 ns
## 4 BL answer repeat 29 29 0.187 28 0.853 1 ns
## 5 BL control meani… 29 29 -1.97 28 0.059 0.587 ns
## 6 BL control meani… 29 29 -0.808 28 0.426 1 ns
## 7 BL control repeat 29 29 -0.297 28 0.769 1 ns
## 8 BL meaningful… meani… 29 29 1.19 28 0.245 1 ns
## 9 BL meaningful… repeat 29 29 0.973 28 0.339 1 ns
## 10 BL meaningles… repeat 29 29 0.132 28 0.896 1 ns## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: BL ~ condition + (1 | sub)
## Data: filter1.on
##
## REML criterion at convergence: 241.1
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.30298 -0.60738 -0.06622 0.60504 3.03104
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.05835 0.2416
## Residual 0.24880 0.4988
## Number of obs: 145, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.32877 0.10291 122.33988 3.195 0.00178 **
## conditioncontrol -0.21822 0.13099 112.00000 -1.666 0.09853 .
## conditionmeaningfully 0.03064 0.13099 112.00000 0.234 0.81547
## conditionmeaninglessly -0.14248 0.13099 112.00000 -1.088 0.27906
## conditionrepeat -0.03751 0.13099 112.00000 -0.286 0.77511
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.636
## cndtnmnngfl -0.636 0.500
## cndtnmnngls -0.636 0.500 0.500
## conditinrpt -0.636 0.500 0.500 0.500## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 1.2541 0.31352 4 112 1.2601 0.29## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 BL answer contr… 29 29 1.92 28 0.065 0.646 ns
## 2 BL answer meani… 29 29 -0.207 28 0.838 1 ns
## 3 BL answer meani… 29 29 1.03 28 0.314 1 ns
## 4 BL answer repeat 29 29 0.333 28 0.741 1 ns
## 5 BL control meani… 29 29 -2.28 28 0.03 0.302 ns
## 6 BL control meani… 29 29 -0.841 28 0.407 1 ns
## 7 BL control repeat 29 29 -1.33 28 0.193 1 ns
## 8 BL meaningful… meani… 29 29 1.27 28 0.213 1 ns
## 9 BL meaningful… repeat 29 29 0.460 28 0.649 1 ns
## 10 BL meaningles… repeat 29 29 -0.650 28 0.521 1 ns3.5 Heterogeneity
Packages for calculating Heterogeneity
if(!require(here)) install.packages("here")## Loading required package: here## here() starts at /Users/betty/Desktoplibrary(here)
if(!require(ggplot2)) install.packages("ggplot2")## Loading required package: ggplot2library(ggplot2)
if(!require(dplyr)) install.packages("dplyr")
library(dplyr)
if(!require(tidyr)) install.packages("tidyr")## Loading required package: tidyr##
## Attaching package: 'tidyr'## The following object is masked from 'package:reshape2':
##
## smiths## The following object is masked from 'package:magrittr':
##
## extract## The following objects are masked from 'package:Matrix':
##
## expand, pack, unpacklibrary(tidyr)
if(!require(viridis)) install.packages("viridis")## Loading required package: viridis## Loading required package: viridisLitelibrary(viridis)
if(!require(gridExtra)) install.packages("gridExtra")## Loading required package: gridExtra##
## Attaching package: 'gridExtra'## The following object is masked from 'package:dplyr':
##
## combinelibrary(gridExtra)
if(!require(readr)) install.packages("readr")## Loading required package: readrlibrary(readr)
if(!require(plyr)) install.packages("plyr")## Loading required package: plyr## ------------------------------------------------------------------------------## You have loaded plyr after dplyr - this is likely to cause problems.
## If you need functions from both plyr and dplyr, please load plyr first, then dplyr:
## library(plyr); library(dplyr)## ------------------------------------------------------------------------------##
## Attaching package: 'plyr'## The following object is masked from 'package:here':
##
## here## The following objects are masked from 'package:dplyr':
##
## arrange, count, desc, failwith, id, mutate, rename, summarise,
## summarize## The following objects are masked from 'package:rstatix':
##
## desc, mutatelibrary(plyr)
if(!require(stringr)) install.packages("stringr")## Loading required package: stringrlibrary(stringr)
if(!require(readxl)) install.packages("readxl")
library(readxl)
if(!require(data.table)) install.packages("data.table")## Loading required package: data.table##
## Attaching package: 'data.table'## The following objects are masked from 'package:reshape2':
##
## dcast, melt## The following objects are masked from 'package:dplyr':
##
## between, first, lastlibrary(data.table)
if(!require(lmerTest)) install.packages("lmerTest")
library(lmerTest)
if(!require(plotly)) install.packages("plotly")## Loading required package: plotly##
## Attaching package: 'plotly'## The following objects are masked from 'package:plyr':
##
## arrange, mutate, rename, summarise## The following object is masked from 'package:ggplot2':
##
## last_plot## The following object is masked from 'package:stats':
##
## filter## The following object is masked from 'package:graphics':
##
## layoutlibrary(plotly)
if(!require(zoo)) install.packages("zoo")## Loading required package: zoo##
## Attaching package: 'zoo'## The following objects are masked from 'package:data.table':
##
## yearmon, yearqtr## The following objects are masked from 'package:base':
##
## as.Date, as.Date.numericlibrary(zoo)
if(!require(viridis)) install.packages("viridis")
library(viridis)
if(!require(brms)) install.packages("brms")## Loading required package: brms## Loading required package: Rcpp## Loading 'brms' package (version 2.22.0). Useful instructions
## can be found by typing help('brms'). A more detailed introduction
## to the package is available through vignette('brms_overview').##
## Attaching package: 'brms'## The following object is masked from 'package:lme4':
##
## ngrps## The following object is masked from 'package:stats':
##
## arlibrary(brms)
if(!require(purrr)) install.packages("purrr")## Loading required package: purrr##
## Attaching package: 'purrr'## The following object is masked from 'package:data.table':
##
## transpose## The following object is masked from 'package:plyr':
##
## compact## The following object is masked from 'package:car':
##
## some## The following object is masked from 'package:magrittr':
##
## set_nameslibrary(purrr)
if(!require(MASS)) install.packages("MASS")## Loading required package: MASS##
## Attaching package: 'MASS'## The following object is masked from 'package:plotly':
##
## select## The following object is masked from 'package:dplyr':
##
## select## The following object is masked from 'package:rstatix':
##
## selectlibrary(MASS)
if(!require(rstan)) install.packages("rstan")## Loading required package: rstan## Loading required package: StanHeaders##
## rstan version 2.32.6 (Stan version 2.32.2)## For execution on a local, multicore CPU with excess RAM we recommend calling
## options(mc.cores = parallel::detectCores()).
## To avoid recompilation of unchanged Stan programs, we recommend calling
## rstan_options(auto_write = TRUE)
## For within-chain threading using `reduce_sum()` or `map_rect()` Stan functions,
## change `threads_per_chain` option:
## rstan_options(threads_per_chain = 1)##
## Attaching package: 'rstan'## The following object is masked from 'package:tidyr':
##
## extract## The following object is masked from 'package:magrittr':
##
## extractlibrary(rstan)
if(!require(ggdist)) install.packages("ggdist")## Loading required package: ggdist##
## Attaching package: 'ggdist'## The following objects are masked from 'package:brms':
##
## dstudent_t, pstudent_t, qstudent_t, rstudent_tlibrary(ggdist)
if(!require(bayestestR)) install.packages("bayestestR")## Loading required package: bayestestR##
## Attaching package: 'bayestestR'## The following object is masked from 'package:ggdist':
##
## hdilibrary(bayestestR)
if(!require(posterior)) install.packages("posterior")## Loading required package: posterior## This is posterior version 1.6.0##
## Attaching package: 'posterior'## The following objects are masked from 'package:rstan':
##
## ess_bulk, ess_tail## The following objects are masked from 'package:stats':
##
## mad, sd, var## The following objects are masked from 'package:base':
##
## %in%, matchlibrary(posterior)
if(!require(distributional)) install.packages("distributional")## Loading required package: distributionallibrary(distributional)
if(!require(cowplot)) install.packages("cowplot")## Loading required package: cowplotlibrary(cowplot)
if(!require(modelr)) install.packages("modelr")## Loading required package: modelrlibrary(modelr)
if(!require(purrr)) install.packages("purrr")
library(purrr)
if(!require(forcats)) install.packages("forcats")## Loading required package: forcatslibrary(forcats)
if(!require(tidybayes)) install.packages("tidybayes")## Loading required package: tidybayes##
## Attaching package: 'tidybayes'## The following object is masked from 'package:distributional':
##
## parameters## The following object is masked from 'package:bayestestR':
##
## hdi## The following objects are masked from 'package:brms':
##
## dstudent_t, pstudent_t, qstudent_t, rstudent_tlibrary(tidybayes)
if(!require(bayesplot)) install.packages("bayesplot")## Loading required package: bayesplot## This is bayesplot version 1.11.1## - Online documentation and vignettes at mc-stan.org/bayesplot## - bayesplot theme set to bayesplot::theme_default()## * Does _not_ affect other ggplot2 plots## * See ?bayesplot_theme_set for details on theme setting##
## Attaching package: 'bayesplot'## The following object is masked from 'package:posterior':
##
## rhat## The following object is masked from 'package:brms':
##
## rhatlibrary(bayesplot)
if(!require(BayesFactor)) install.packages("BayesFactor")## Loading required package: BayesFactor## Loading required package: coda##
## Attaching package: 'coda'## The following object is masked from 'package:rstan':
##
## traceplot## ************
## Welcome to BayesFactor 0.9.12-4.7. If you have questions, please contact Richard Morey (richarddmorey@gmail.com).
##
## Type BFManual() to open the manual.
## ************library(BayesFactor)
if(!require(patchwork)) install.packages("patchwork")## Loading required package: patchwork##
## Attaching package: 'patchwork'## The following object is masked from 'package:cowplot':
##
## align_plots## The following object is masked from 'package:MASS':
##
## arealibrary(patchwork)
if(!require(scales)) install.packages("scales")## Loading required package: scales##
## Attaching package: 'scales'## The following object is masked from 'package:purrr':
##
## discard## The following object is masked from 'package:readr':
##
## col_factor## The following object is masked from 'package:viridis':
##
## viridis_pallibrary(scales)SDLP
# scale the data
SDLPdata$age.c <- scale(SDLPdata$age, center = T, scale = F)
# fit model
Model <- lmer(data = SDLPdata, SDLP~beforeafter*condition*age.c+(1+beforeafter|sub))## boundary (singular) fit: see help('isSingular')summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: SDLP ~ beforeafter * condition * age.c + (1 + beforeafter | sub)
## Data: SDLPdata
##
## REML criterion at convergence: 294
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -4.0678 -0.4927 -0.0083 0.5983 2.7487
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## sub (Intercept) 0.07328 0.27071
## beforeafteron 0.00361 0.06008 -1.00
## Residual 0.09357 0.30589
## Number of obs: 290, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df
## (Intercept) -0.071573 0.075852 82.622088
## beforeafteron -0.177340 0.081101 235.165557
## conditioncontrol -0.123441 0.080330 242.999694
## conditionmeaningfully -0.040493 0.080330 243.002196
## conditionmeaninglessly -0.045602 0.080330 242.999694
## conditionrepeat 0.114622 0.080330 242.999694
## age.c -0.002878 0.007417 82.684992
## beforeafteron:conditioncontrol 0.346896 0.113604 242.999694
## beforeafteron:conditionmeaningfully 0.253119 0.113604 242.999756
## beforeafteron:conditionmeaninglessly 0.332040 0.113604 242.999694
## beforeafteron:conditionrepeat -0.131589 0.113604 242.999694
## beforeafteron:age.c 0.011129 0.007931 235.171190
## conditioncontrol:age.c 0.004205 0.007855 242.999694
## conditionmeaningfully:age.c 0.001761 0.007851 243.001331
## conditionmeaninglessly:age.c 0.009461 0.007855 242.999694
## conditionrepeat:age.c 0.002628 0.007855 242.999694
## beforeafteron:conditioncontrol:age.c -0.003036 0.011109 242.999694
## beforeafteron:conditionmeaningfully:age.c -0.013130 0.011103 242.999735
## beforeafteron:conditionmeaninglessly:age.c -0.016123 0.011109 242.999694
## beforeafteron:conditionrepeat:age.c -0.008029 0.011109 242.999694
## t value Pr(>|t|)
## (Intercept) -0.944 0.34813
## beforeafteron -2.187 0.02975 *
## conditioncontrol -1.537 0.12567
## conditionmeaningfully -0.504 0.61466
## conditionmeaninglessly -0.568 0.57077
## conditionrepeat 1.427 0.15490
## age.c -0.388 0.69902
## beforeafteron:conditioncontrol 3.054 0.00251 **
## beforeafteron:conditionmeaningfully 2.228 0.02679 *
## beforeafteron:conditionmeaninglessly 2.923 0.00380 **
## beforeafteron:conditionrepeat -1.158 0.24787
## beforeafteron:age.c 1.403 0.16184
## conditioncontrol:age.c 0.535 0.59295
## conditionmeaningfully:age.c 0.224 0.82269
## conditionmeaninglessly:age.c 1.204 0.22958
## conditionrepeat:age.c 0.335 0.73828
## beforeafteron:conditioncontrol:age.c -0.273 0.78487
## beforeafteron:conditionmeaningfully:age.c -1.183 0.23813
## beforeafteron:conditionmeaninglessly:age.c -1.451 0.14797
## beforeafteron:conditionrepeat:age.c -0.723 0.47051
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Correlation matrix not shown by default, as p = 20 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')# calculate the contribution of age
# Get interaction coefficient between age and beforeafter
bXa.coeff <- fixef(Model)[12]
# Random effect in model for beforeafter - squared to compute variance
residvalvar <- 0.00361
# Calculate the implied total random effect variance - interaction squared multiplied by variance in age + residual variance.
imptotalvalvar <- bXa.coeff^2*sd(SDLPdata$age.c)^2 + residvalvar
# Variance in mu difference explained by age
V_SDLP <-1- (residvalvar/imptotalvalvar)
V_SDLP## beforeafteron:age.c
## 0.7827057# plot the data
## computing random effects
ageranef_original <- as.data.frame(ranef(Model)) %>%
dplyr::select("grp","term","condval")
library(reshape2)
ageranef <- dcast(ageranef_original,
grp~term,
timevar = c("condval")) ## Warning in dcast(ageranef_original, grp ~ term, timevar = c("condval")): The
## dcast generic in data.table has been passed a data.frame and will attempt to
## redirect to the relevant reshape2 method; please note that reshape2 is
## superseded and is no longer actively developed, and this redirection is now
## deprecated. Please do this redirection yourself like
## reshape2::dcast(ageranef_original). In the next version, this warning will
## become an error.## Using 'condval' as value column. Use 'value.var' to overrideageranef <- ageranef %>% rename("intercept_age"="(Intercept)","slope_age"="beforeafteron","sub"="grp")
## Create dataset with one line per person with mileage score
ageranef <- merge(ageranef, SDLPdata, by = 'sub')
ageranef <- ageranef %>%
dplyr::group_by(sub) %>%
dplyr::slice(1) %>%
dplyr::select(sub, age.c, intercept_age, slope_age)
## Person-specific implied total n-back effects for model accounting for age
ranef.age.pred.tot <- fixef(Model)[2] + # fixed effect for n-back
fixef(Model)[12] * ageranef$age.c + # fixed effect for n-back X age interaction # scale of age
ageranef$slope_age # estimate random effect of slopes
## relationship between the effect of N-back and age
age.nback.effect.sge_SDLP <- ggplot(ageranef, aes(age.c, ranef.age.pred.tot)) +
geom_vline(xintercept = 0, size = .5, color = "black", linetype="solid") +
geom_hline(yintercept = mean(ranef.age.pred.tot), size = .5, color = "black", linetype="solid") +
geom_jitter(height = 0, width = 0, size = 4,
shape = 21, colour = "black", fill = viridis(5)[3], alpha = .95, stroke = 1) +
stat_smooth(method = 'lm', ci = T, color = viridis(5)[3], size = 1) +
xlab("Age (mean centered)") +
ylab(expression("Implied total heterogeneity of "~ SDLP)) +
#ylim(-.15, .1) +
theme_bw() +
theme(legend.position = "bottom", legend.direction = "horizontal", axis.title.x = element_text(size = 11), axis.text.x = element_text(size = 15), axis.title.y = element_text(size = 11), axis.text.y = element_text(size = 15), title = element_text(size = 18), legend.title = element_text(size = 12), legend.text = element_text(size = 12), legend.key = element_blank(), legend.key.width = unit(0.3, 'cm'), legend.key.size = unit(0.1, 'cm'), plot.title = element_text(size = 7.5, face = "bold"), strip.text = element_text(face = "bold", size = 10))## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.## Warning in stat_smooth(method = "lm", ci = T, color = viridis(5)[3], size = 1):
## Ignoring unknown parameters: `ci`age.nback.effect.sge_SDLP## `geom_smooth()` using formula = 'y ~ x'
SDVHA
# scale the data
SDVHdata$age.c <- scale(SDVHdata$age, center = T, scale = F)
# fit model
Model <- lmer(data = SDVHdata, SDVH~beforeafter*condition*age.c+(1+beforeafter|sub))## boundary (singular) fit: see help('isSingular')summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: SDVH ~ beforeafter * condition * age.c + (1 + beforeafter | sub)
## Data: SDVHdata
##
## REML criterion at convergence: 424
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -7.7957 -0.4702 0.0008 0.5295 2.0262
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## sub (Intercept) 0.1126592 0.33565
## beforeafteron 0.0003094 0.01759 1.00
## Residual 0.1482212 0.38500
## Number of obs: 290, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df
## (Intercept) -4.073e-01 9.485e-02 8.359e+01
## beforeafteron -2.748e-01 1.012e-01 2.425e+02
## conditioncontrol -1.471e-01 1.011e-01 2.430e+02
## conditionmeaningfully -1.058e-01 1.011e-01 2.430e+02
## conditionmeaninglessly -1.130e-01 1.011e-01 2.430e+02
## conditionrepeat 1.097e-01 1.011e-01 2.430e+02
## age.c 3.113e-04 9.274e-03 8.366e+01
## beforeafteron:conditioncontrol 4.253e-01 1.430e-01 2.430e+02
## beforeafteron:conditionmeaningfully 2.809e-01 1.430e-01 2.430e+02
## beforeafteron:conditionmeaninglessly 4.853e-01 1.430e-01 2.430e+02
## beforeafteron:conditionrepeat -1.937e-01 1.430e-01 2.430e+02
## beforeafteron:age.c 1.209e-02 9.892e-03 2.426e+02
## conditioncontrol:age.c 7.923e-03 9.887e-03 2.430e+02
## conditionmeaningfully:age.c -5.967e-03 9.881e-03 2.430e+02
## conditionmeaninglessly:age.c 5.817e-03 9.887e-03 2.430e+02
## conditionrepeat:age.c -3.506e-03 9.887e-03 2.430e+02
## beforeafteron:conditioncontrol:age.c -7.788e-03 1.398e-02 2.430e+02
## beforeafteron:conditionmeaningfully:age.c -7.995e-03 1.397e-02 2.430e+02
## beforeafteron:conditionmeaninglessly:age.c -1.822e-02 1.398e-02 2.430e+02
## beforeafteron:conditionrepeat:age.c -3.953e-04 1.398e-02 2.430e+02
## t value Pr(>|t|)
## (Intercept) -4.295 4.69e-05 ***
## beforeafteron -2.716 0.007081 **
## conditioncontrol -1.455 0.147034
## conditionmeaningfully -1.046 0.296427
## conditionmeaninglessly -1.118 0.264803
## conditionrepeat 1.085 0.279175
## age.c 0.034 0.973306
## beforeafteron:conditioncontrol 2.975 0.003228 **
## beforeafteron:conditionmeaningfully 1.965 0.050610 .
## beforeafteron:conditionmeaninglessly 3.394 0.000803 ***
## beforeafteron:conditionrepeat -1.354 0.176859
## beforeafteron:age.c 1.222 0.222905
## conditioncontrol:age.c 0.801 0.423726
## conditionmeaningfully:age.c -0.604 0.546510
## conditionmeaninglessly:age.c 0.588 0.556810
## conditionrepeat:age.c -0.355 0.723202
## beforeafteron:conditioncontrol:age.c -0.557 0.578050
## beforeafteron:conditionmeaningfully:age.c -0.572 0.567761
## beforeafteron:conditionmeaninglessly:age.c -1.303 0.193692
## beforeafteron:conditionrepeat:age.c -0.028 0.977468
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Correlation matrix not shown by default, as p = 20 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')# calculate the contribution of age
# Get interaction coefficient between age and beforeafter
bXa.coeff <- fixef(Model)[12]
# Random effect in model for beforeafter - squared to compute variance
residvalvar <- 0.0003094
# Calculate the implied total random effect variance - interaction squared multiplied by variance in age + residual variance.
imptotalvalvar <- bXa.coeff^2*sd(SDVHdata$age.c)^2 + residvalvar
# Variance in mu difference explained by age
V_SDVH <- 1 - (residvalvar/imptotalvalvar)
V_SDVH## beforeafteron:age.c
## 0.9802287# plot the data
## computing random effects
ageranef_original <- as.data.frame(ranef(Model)) %>%
dplyr::select("grp","term","condval")
library(reshape2)
ageranef <- dcast(ageranef_original,
grp~term,
timevar = c("condval")) ## Warning in dcast(ageranef_original, grp ~ term, timevar = c("condval")): The
## dcast generic in data.table has been passed a data.frame and will attempt to
## redirect to the relevant reshape2 method; please note that reshape2 is
## superseded and is no longer actively developed, and this redirection is now
## deprecated. Please do this redirection yourself like
## reshape2::dcast(ageranef_original). In the next version, this warning will
## become an error.## Using 'condval' as value column. Use 'value.var' to overrideageranef <- ageranef %>% rename("intercept_age"="(Intercept)","slope_age"="beforeafteron","sub"="grp")
#Create dataset with one line per person with mileage score
ageranef <- merge(ageranef, SDVHdata, by = 'sub')
ageranef <- ageranef %>%
dplyr::group_by(sub) %>%
dplyr::slice(1) %>%
dplyr::select(sub, age.c, intercept_age, slope_age)
# Person-specific implied total n-back effects for model accounting for age
ranef.age.pred.tot <- fixef(Model)[2] + # fixed effect for n-back
fixef(Model)[12] * ageranef$age.c + # fixed effect for n-back X age interaction # scale of age
ageranef$slope_age # estimate random effect of slopes
# relationship between the effect of N-back and age
age.nback.effect.sge_SDVH <- ggplot(ageranef, aes(age.c, ranef.age.pred.tot)) +
geom_vline(xintercept = 0, size = .5, color = "black", linetype="solid") +
geom_hline(yintercept = mean(ranef.age.pred.tot), size = .5, color = "black", linetype="solid") +
geom_jitter(height = 0, width = 0, size = 4,
shape = 21, colour = "black", fill = viridis(5)[3], alpha = .95, stroke = 1) +
stat_smooth(method = 'lm', ci = T, color = viridis(5)[3], size = 1) +
xlab("Age (mean centered)") +
ylab(expression("Implied total heterogeneity of "~ SDVH)) +
#ylim(-.15, .1) +
theme_bw() +
theme(legend.position = "bottom", legend.direction = "horizontal", axis.title.x = element_text(size = 11), axis.text.x = element_text(size = 15), axis.title.y = element_text(size = 11), axis.text.y = element_text(size = 15), title = element_text(size = 18), legend.title = element_text(size = 12), legend.text = element_text(size = 12), legend.key = element_blank(), legend.key.width = unit(0.3, 'cm'), legend.key.size = unit(0.1, 'cm'), plot.title = element_text(size = 7.5, face = "bold"), strip.text = element_text(face = "bold", size = 10))## Warning in stat_smooth(method = "lm", ci = T, color = viridis(5)[3], size = 1):
## Ignoring unknown parameters: `ci`age.nback.effect.sge_SDVH## `geom_smooth()` using formula = 'y ~ x'
SDspeed
# scale the data
SDSpeeddata$age.c <- scale(SDSpeeddata$age, center = T, scale = F)
# fit model
Model <- lmer(data = SDSpeeddata, SDSpeed~beforeafter*condition*age.c+(1+beforeafter|sub))## boundary (singular) fit: see help('isSingular')summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: SDSpeed ~ beforeafter * condition * age.c + (1 + beforeafter |
## sub)
## Data: SDSpeeddata
##
## REML criterion at convergence: 817.2
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -4.5896 -0.4541 0.0747 0.5979 1.9285
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## sub (Intercept) 0.24628 0.4963
## beforeafteron 0.02416 0.1554 -1.00
## Residual 0.69695 0.8348
## Number of obs: 290, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df
## (Intercept) 0.628037 0.180348 130.276714
## beforeafteron 0.753303 0.221131 235.454737
## conditioncontrol 0.163893 0.219239 242.999730
## conditionmeaningfully 0.166273 0.219240 243.001079
## conditionmeaninglessly 0.490918 0.219239 242.999730
## conditionrepeat 0.137747 0.219239 242.999730
## age.c 0.005288 0.017636 130.312298
## beforeafteron:conditioncontrol -0.516859 0.310051 242.999730
## beforeafteron:conditionmeaningfully -0.075008 0.310052 242.999796
## beforeafteron:conditionmeaninglessly -0.514545 0.310051 242.999730
## beforeafteron:conditionrepeat -0.060109 0.310051 242.999730
## beforeafteron:age.c -0.001357 0.021624 235.457148
## conditioncontrol:age.c 0.049582 0.021439 242.999730
## conditionmeaningfully:age.c -0.023625 0.021427 243.002807
## conditionmeaninglessly:age.c -0.001845 0.021439 242.999730
## conditionrepeat:age.c -0.013558 0.021439 242.999730
## beforeafteron:conditioncontrol:age.c -0.017962 0.030319 242.999730
## beforeafteron:conditionmeaningfully:age.c 0.017549 0.030302 242.999881
## beforeafteron:conditionmeaninglessly:age.c -0.011106 0.030319 242.999730
## beforeafteron:conditionrepeat:age.c -0.001470 0.030319 242.999730
## t value Pr(>|t|)
## (Intercept) 3.482 0.000677 ***
## beforeafteron 3.407 0.000773 ***
## conditioncontrol 0.748 0.455453
## conditionmeaningfully 0.758 0.448942
## conditionmeaninglessly 2.239 0.026049 *
## conditionrepeat 0.628 0.530400
## age.c 0.300 0.764770
## beforeafteron:conditioncontrol -1.667 0.096800 .
## beforeafteron:conditionmeaningfully -0.242 0.809044
## beforeafteron:conditionmeaninglessly -1.660 0.098295 .
## beforeafteron:conditionrepeat -0.194 0.846441
## beforeafteron:age.c -0.063 0.950026
## conditioncontrol:age.c 2.313 0.021576 *
## conditionmeaningfully:age.c -1.103 0.271314
## conditionmeaninglessly:age.c -0.086 0.931500
## conditionrepeat:age.c -0.632 0.527713
## beforeafteron:conditioncontrol:age.c -0.592 0.554121
## beforeafteron:conditionmeaningfully:age.c 0.579 0.563040
## beforeafteron:conditionmeaninglessly:age.c -0.366 0.714454
## beforeafteron:conditionrepeat:age.c -0.048 0.961383
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Correlation matrix not shown by default, as p = 20 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')# calculate the contribution of age
# Get interaction coefficient between age and beforeafter
bXa.coeff <- fixef(Model)[12]
# Random effect in model for beforeafter - squared to compute variance
residvalvar <- 0.02416
# Calculate the implied total random effect variance - interaction squared multiplied by variance in age + residual variance.
imptotalvalvar <- bXa.coeff^2*sd(SDSpeeddata$age.c)^2 + residvalvar
# Variance in mu difference explained by age
V_SD_speed<- 1 - (residvalvar/imptotalvalvar)
V_SD_speed## beforeafteron:age.c
## 0.007934721# plot the data
## computing random effects
ageranef_original <- as.data.frame(ranef(Model)) %>%
dplyr::select("grp","term","condval")
library(reshape2)
ageranef <- dcast(ageranef_original,
grp~term,
timevar = c("condval")) ## Warning in dcast(ageranef_original, grp ~ term, timevar = c("condval")): The
## dcast generic in data.table has been passed a data.frame and will attempt to
## redirect to the relevant reshape2 method; please note that reshape2 is
## superseded and is no longer actively developed, and this redirection is now
## deprecated. Please do this redirection yourself like
## reshape2::dcast(ageranef_original). In the next version, this warning will
## become an error.## Using 'condval' as value column. Use 'value.var' to overrideageranef <- ageranef %>% rename("intercept_age"="(Intercept)","slope_age"="beforeafteron","sub"="grp")
## Create dataset with one line per person with mileage score
ageranef <- merge(ageranef, SDSpeeddata, by = 'sub')
ageranef <- ageranef %>%
dplyr::group_by(sub) %>%
dplyr::slice(1) %>%
dplyr::select(sub, age.c, intercept_age, slope_age)
## Person-specific implied total n-back effects for model accounting for age
ranef.age.pred.tot <- fixef(Model)[2] + # fixed effect for n-back
fixef(Model)[12] * ageranef$age.c + # fixed effect for n-back X age interaction # scale of age
ageranef$slope_age # estimate random effect of slopes
## relationship between the effect of N-back and age
age.nback.effect.sge_SDSpeed <- ggplot(ageranef, aes(age.c, ranef.age.pred.tot)) +
geom_vline(xintercept = 0, size = .5, color = "black", linetype="solid") +
geom_hline(yintercept = mean(ranef.age.pred.tot), size = .5, color = "black", linetype="solid") +
geom_jitter(height = 0, width = 0, size = 4,
shape = 21, colour = "black", fill = viridis(5)[3], alpha = .95, stroke = 1) +
stat_smooth(method = 'lm', ci = T, color = viridis(5)[3], size = 1) +
xlab("Age (mean centered)") +
ylab(expression("Implied total heterogeneity of "~ SD-speed)) +
#ylim(-.15, .1) +
theme_bw() +
theme(legend.position = "bottom", legend.direction = "horizontal", axis.title.x = element_text(size = 11), axis.text.x = element_text(size = 15), axis.title.y = element_text(size = 11), axis.text.y = element_text(size = 15), title = element_text(size = 18), legend.title = element_text(size = 12), legend.text = element_text(size = 12), legend.key = element_blank(), legend.key.width = unit(0.3, 'cm'), legend.key.size = unit(0.1, 'cm'), plot.title = element_text(size = 7.5, face = "bold"), strip.text = element_text(face = "bold", size = 10))## Warning in stat_smooth(method = "lm", ci = T, color = viridis(5)[3], size = 1):
## Ignoring unknown parameters: `ci`age.nback.effect.sge_SDSpeed## `geom_smooth()` using formula = 'y ~ x'
SCL
# scale the data
SCLdata$age.c <- scale(SCLdata$age, center = T, scale = F)
# fit model
Model <- lmer(data = SCLdata, SCL~beforeafter*condition+(1+beforeafter|sub))## boundary (singular) fit: see help('isSingular')## Warning: Model failed to converge with 1 negative eigenvalue: -9.8e+01summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: SCL ~ beforeafter * condition + (1 + beforeafter | sub)
## Data: SCLdata
##
## REML criterion at convergence: 553.9
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.2647 -0.5144 -0.1277 0.2994 5.1168
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## sub (Intercept) 0.00000 0.0000
## beforeafteron 0.05386 0.2321 NaN
## Residual 0.35476 0.5956
## Number of obs: 290, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df t value
## (Intercept) 1.16091 0.11060 248.07117 10.496
## beforeafteron 0.82918 0.16225 255.04038 5.111
## conditioncontrol -0.06518 0.15642 248.07117 -0.417
## conditionmeaningfully 0.10921 0.15642 248.07117 0.698
## conditionmeaninglessly -0.13121 0.15642 248.07117 -0.839
## conditionrepeat -0.07328 0.15642 248.07117 -0.468
## beforeafteron:conditioncontrol -0.90331 0.22121 248.07117 -4.084
## beforeafteron:conditionmeaningfully -0.45642 0.22121 248.07117 -2.063
## beforeafteron:conditionmeaninglessly -0.71217 0.22121 248.07117 -3.219
## beforeafteron:conditionrepeat -0.11845 0.22121 248.07117 -0.535
## Pr(>|t|)
## (Intercept) < 2e-16 ***
## beforeafteron 6.30e-07 ***
## conditioncontrol 0.67727
## conditionmeaningfully 0.48573
## conditionmeaninglessly 0.40237
## conditionrepeat 0.63987
## beforeafteron:conditioncontrol 5.99e-05 ***
## beforeafteron:conditionmeaningfully 0.04012 *
## beforeafteron:conditionmeaninglessly 0.00146 **
## beforeafteron:conditionrepeat 0.59282
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) bfrftr cndtnc cndtnmnngf cndtnmnngl cndtnr
## beforeaftrn -0.682
## condtncntrl -0.707 0.482
## cndtnmnngfl -0.707 0.482 0.500
## cndtnmnngls -0.707 0.482 0.500 0.500
## conditinrpt -0.707 0.482 0.500 0.500 0.500
## bfrftrn:cndtnc 0.500 -0.682 -0.707 -0.354 -0.354 -0.354
## bfrftrn:cndtnmnngf 0.500 -0.682 -0.354 -0.707 -0.354 -0.354
## bfrftrn:cndtnmnngl 0.500 -0.682 -0.354 -0.354 -0.707 -0.354
## bfrftrn:cndtnr 0.500 -0.682 -0.354 -0.354 -0.354 -0.707
## bfrftrn:cndtnc bfrftrn:cndtnmnngf bfrftrn:cndtnmnngl
## beforeaftrn
## condtncntrl
## cndtnmnngfl
## cndtnmnngls
## conditinrpt
## bfrftrn:cndtnc
## bfrftrn:cndtnmnngf 0.500
## bfrftrn:cndtnmnngl 0.500 0.500
## bfrftrn:cndtnr 0.500 0.500 0.500
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')# calculate the contribution of age
# Get interaction coefficient between age and beforeafter
bXa.coeff <- fixef(Model)[12] # the estimate parameter of age.c and beforeafter interaction effect
#Random effect in model for beforeafter - squared to compute variance
residvalvar <- 0.05661
#Calculate the implied total random effect variance - interaction squared multiplied by variance in age + residual variance.
imptotalvalvar <- bXa.coeff^2*sd(SCLdata$age.c)^2 + residvalvar
## Variance in mu difference explained by age
V_SCL <- 1 - (residvalvar/imptotalvalvar)
V_SCL## <NA>
## NA# plot the data
# computing random effects
ageranef_original <- as.data.frame(ranef(Model)) %>%
dplyr::select("grp","term","condval")
library(reshape2)
ageranef <- dcast(ageranef_original,
grp~term,
timevar = c("condval")) ## Warning in dcast(ageranef_original, grp ~ term, timevar = c("condval")): The
## dcast generic in data.table has been passed a data.frame and will attempt to
## redirect to the relevant reshape2 method; please note that reshape2 is
## superseded and is no longer actively developed, and this redirection is now
## deprecated. Please do this redirection yourself like
## reshape2::dcast(ageranef_original). In the next version, this warning will
## become an error.## Using 'condval' as value column. Use 'value.var' to overrideageranef <- ageranef %>% rename("intercept_age"="(Intercept)","slope_age"="beforeafteron","sub"="grp")
#Create dataset with one line per person with mileage score
ageranef <- merge(ageranef, SCLdata, by = 'sub')
ageranef <- ageranef %>%
dplyr::group_by(sub) %>%
dplyr::slice(1) %>%
dplyr::select(sub, age.c, intercept_age, slope_age)
# Person-specific implied total n-back effects for model accounting for age
ranef.age.pred.tot <- fixef(Model)[2] + # fixed effect for n-back
fixef(Model)[12] * ageranef$age.c + # fixed effect for n-back X age interaction # scale of age
ageranef$slope_age # estimate random effect of slopes
# relationship between the effect of N-back and age
age.nback.effect.sge_SCL <- ggplot(ageranef, aes(age.c, ranef.age.pred.tot)) +
geom_vline(xintercept = 0, size = .5, color = "black", linetype="solid") +
geom_hline(yintercept = mean(ranef.age.pred.tot), size = .5, color = "black", linetype="solid") +
geom_jitter(height = 0, width = 0, size = 4,
shape = 21, colour = "black", fill = viridis(5)[3], alpha = .95, stroke = 1) +
stat_smooth(method = 'lm', ci = T, color = viridis(5)[3], size = 1) +
xlab("Age (mean centered)") +
ylab(expression("Implied total heterogeneity of "~ SCL)) +
#ylim(-.15, .1) +
theme_bw() +
theme(legend.position = "bottom", legend.direction = "horizontal", axis.title.x = element_text(size = 11), axis.text.x = element_text(size = 15), axis.title.y = element_text(size = 11), axis.text.y = element_text(size = 15), title = element_text(size = 18), legend.title = element_text(size = 12), legend.text = element_text(size = 12), legend.key = element_blank(), legend.key.width = unit(0.3, 'cm'), legend.key.size = unit(0.1, 'cm'), plot.title = element_text(size = 7.5, face = "bold"), strip.text = element_text(face = "bold", size = 10))## Warning in stat_smooth(method = "lm", ci = T, color = viridis(5)[3], size = 1):
## Ignoring unknown parameters: `ci`age.nback.effect.sge_SCL## `geom_smooth()` using formula = 'y ~ x'## Warning: Removed 29 rows containing non-finite outside the scale range
## (`stat_smooth()`).## Warning: Removed 1 row containing missing values or values outside the scale range
## (`geom_hline()`).## Warning: Removed 29 rows containing missing values or values outside the scale range
## (`geom_point()`).
SDNN
# scale the data
SDNNdata$age.c <- scale(SDNNdata$age, center = T, scale = F)
# fit model
library(lmerTest)
Model <- lmer(data = SDNNdata, SDNN~beforeafter*condition*age.c+(1+beforeafter|sub))## boundary (singular) fit: see help('isSingular')## Warning: Model failed to converge with 1 negative eigenvalue: -2.6e+02summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: SDNN ~ beforeafter * condition * age.c + (1 + beforeafter | sub)
## Data: SDNNdata
##
## REML criterion at convergence: 321.9
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.2872 -0.6244 -0.0278 0.5586 2.6948
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## sub (Intercept) 0.00000 0.0000
## beforeafteron 0.01036 0.1018 NaN
## Residual 0.12212 0.3495
## Number of obs: 290, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df
## (Intercept) -4.755e-02 6.489e-02 2.419e+02
## beforeafteron 3.365e-01 9.370e-02 2.644e+02
## conditioncontrol 1.608e-01 9.177e-02 2.419e+02
## conditionmeaningfully 1.311e-01 9.177e-02 2.419e+02
## conditionmeaninglessly 2.217e-01 9.177e-02 2.419e+02
## conditionrepeat 1.981e-02 9.177e-02 2.419e+02
## age.c -1.721e-03 6.346e-03 2.419e+02
## beforeafteron:conditioncontrol -4.848e-01 1.298e-01 2.419e+02
## beforeafteron:conditionmeaningfully -2.026e-01 1.298e-01 2.419e+02
## beforeafteron:conditionmeaninglessly -2.703e-01 1.298e-01 2.419e+02
## beforeafteron:conditionrepeat 1.097e-01 1.298e-01 2.419e+02
## beforeafteron:age.c 4.602e-03 9.162e-03 2.644e+02
## conditioncontrol:age.c 1.237e-03 8.974e-03 2.419e+02
## conditionmeaningfully:age.c -4.862e-03 8.969e-03 2.419e+02
## conditionmeaninglessly:age.c 1.990e-03 8.974e-03 2.419e+02
## conditionrepeat:age.c 7.110e-03 8.974e-03 2.419e+02
## beforeafteron:conditioncontrol:age.c -7.075e-04 1.269e-02 2.419e+02
## beforeafteron:conditionmeaningfully:age.c -4.887e-03 1.268e-02 2.419e+02
## beforeafteron:conditionmeaninglessly:age.c -1.107e-02 1.269e-02 2.419e+02
## beforeafteron:conditionrepeat:age.c -1.284e-02 1.269e-02 2.419e+02
## t value Pr(>|t|)
## (Intercept) -0.733 0.464377
## beforeafteron 3.592 0.000391 ***
## conditioncontrol 1.752 0.081039 .
## conditionmeaningfully 1.429 0.154425
## conditionmeaninglessly 2.415 0.016461 *
## conditionrepeat 0.216 0.829285
## age.c -0.271 0.786408
## beforeafteron:conditioncontrol -3.735 0.000234 ***
## beforeafteron:conditionmeaningfully -1.561 0.119907
## beforeafteron:conditionmeaninglessly -2.083 0.038337 *
## beforeafteron:conditionrepeat 0.845 0.398711
## beforeafteron:age.c 0.502 0.615908
## conditioncontrol:age.c 0.138 0.890465
## conditionmeaningfully:age.c -0.542 0.588272
## conditionmeaninglessly:age.c 0.222 0.824731
## conditionrepeat:age.c 0.792 0.428966
## beforeafteron:conditioncontrol:age.c -0.056 0.955588
## beforeafteron:conditionmeaningfully:age.c -0.385 0.700358
## beforeafteron:conditionmeaninglessly:age.c -0.872 0.384070
## beforeafteron:conditionrepeat:age.c -1.012 0.312665
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Correlation matrix not shown by default, as p = 20 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')### calculate the contribution of age
# Get interaction coefficient between age and beforeafter
bXa.coeff <- fixef(Model)[12] # the estimate parameter of age.c and beforeafter interaction effect
#Random effect in model for beforeafter - squared to compute variance
residvalvar <- 0.01036
#Calculate the implied total random effect variance - interaction squared multiplied by variance in age + residual variance.
imptotalvalvar <- bXa.coeff^2*sd(SDNNdata$age.c)^2 + residvalvar
## Variance in mu difference explained by age
V_SDNN <- 1 - (residvalvar/imptotalvalvar)
V_SDNN## beforeafteron:age.c
## 0.1766829# plot the data
## computing random effects
ageranef_original <- as.data.frame(ranef(Model)) %>%
dplyr::select("grp","term","condval")
library(reshape2)
ageranef <- dcast(ageranef_original,
grp~term,
timevar = c("condval")) ## Warning in dcast(ageranef_original, grp ~ term, timevar = c("condval")): The
## dcast generic in data.table has been passed a data.frame and will attempt to
## redirect to the relevant reshape2 method; please note that reshape2 is
## superseded and is no longer actively developed, and this redirection is now
## deprecated. Please do this redirection yourself like
## reshape2::dcast(ageranef_original). In the next version, this warning will
## become an error.## Using 'condval' as value column. Use 'value.var' to overrideageranef <- ageranef %>% rename("intercept_age"="(Intercept)","slope_age"="beforeafteron","sub"="grp")
## Create dataset with one line per person with mileage score
ageranef <- merge(ageranef, SDNNdata, by = 'sub')
ageranef <- ageranef %>%
dplyr::group_by(sub) %>%
dplyr::slice(1) %>%
dplyr::select(sub, age.c, intercept_age, slope_age)
## Person-specific implied total n-back effects for model accounting for age
ranef.age.pred.tot <- fixef(Model)[2] + # fixed effect for n-back
fixef(Model)[12] * ageranef$age.c + # fixed effect for n-back X age interaction # scale of age
ageranef$slope_age # estimate random effect of slopes
## relationship between the effect of N-back and age
age.nback.effect.sge_SDNN <- ggplot(ageranef, aes(age.c, ranef.age.pred.tot)) +
geom_vline(xintercept = 0, size = .5, color = "black", linetype="solid") +
geom_hline(yintercept = mean(ranef.age.pred.tot), size = .5, color = "black", linetype="solid") +
geom_jitter(height = 0, width = 0, size = 4,
shape = 21, colour = "black", fill = viridis(5)[3], alpha = .95, stroke = 1) +
stat_smooth(method = 'lm', ci = T, color = viridis(5)[3], size = 1) +
xlab("Age (mean centered)") +
ylab(expression("Implied total heterogeneity of "~ SDNN)) +
#ylim(-.15, .1) +
theme_bw() +
theme(legend.position = "bottom", legend.direction = "horizontal", axis.title.x = element_text(size = 11), axis.text.x = element_text(size = 15), axis.title.y = element_text(size = 11), axis.text.y = element_text(size = 15), title = element_text(size = 18), legend.title = element_text(size = 12), legend.text = element_text(size = 12), legend.key = element_blank(), legend.key.width = unit(0.3, 'cm'), legend.key.size = unit(0.1, 'cm'), plot.title = element_text(size = 7.5, face = "bold"), strip.text = element_text(face = "bold", size = 10))## Warning in stat_smooth(method = "lm", ci = T, color = viridis(5)[3], size = 1):
## Ignoring unknown parameters: `ci`age.nback.effect.sge_SDNN## `geom_smooth()` using formula = 'y ~ x'
Breathing rate
# scale the data
RESPdata$age.c <- scale(RESPdata$age, center = T, scale = F)
# fit model
Model <- lmer(data = RESPdata, RESP~beforeafter*condition*age.c+(1+beforeafter|sub))## boundary (singular) fit: see help('isSingular')summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: RESP ~ beforeafter * condition * age.c + (1 + beforeafter | sub)
## Data: RESPdata
##
## REML criterion at convergence: 189.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -5.5569 -0.3752 0.0764 0.5276 3.0629
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## sub (Intercept) 0.008145 0.09025
## beforeafteron 0.002668 0.05165 1.00
## Residual 0.069295 0.26324
## Number of obs: 290, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df
## (Intercept) -1.208e-01 5.168e-02 1.998e+02
## beforeafteron 4.558e-02 6.979e-02 2.367e+02
## conditioncontrol 1.272e-01 6.913e-02 2.430e+02
## conditionmeaningfully 1.247e-01 6.913e-02 2.430e+02
## conditionmeaninglessly 1.057e-01 6.913e-02 2.430e+02
## conditionrepeat 1.597e-01 6.913e-02 2.430e+02
## age.c -8.163e-04 5.053e-03 1.998e+02
## beforeafteron:conditioncontrol -6.163e-02 9.777e-02 2.430e+02
## beforeafteron:conditionmeaningfully -1.489e-01 9.777e-02 2.430e+02
## beforeafteron:conditionmeaninglessly -1.058e-01 9.777e-02 2.430e+02
## beforeafteron:conditionrepeat -5.216e-02 9.777e-02 2.430e+02
## beforeafteron:age.c 3.779e-03 6.825e-03 2.367e+02
## conditioncontrol:age.c 1.945e-04 6.760e-03 2.430e+02
## conditionmeaningfully:age.c -6.917e-04 6.756e-03 2.430e+02
## conditionmeaninglessly:age.c -4.107e-03 6.760e-03 2.430e+02
## conditionrepeat:age.c 2.752e-04 6.760e-03 2.430e+02
## beforeafteron:conditioncontrol:age.c -4.584e-03 9.560e-03 2.430e+02
## beforeafteron:conditionmeaningfully:age.c 4.232e-03 9.555e-03 2.430e+02
## beforeafteron:conditionmeaninglessly:age.c 1.451e-03 9.560e-03 2.430e+02
## beforeafteron:conditionrepeat:age.c -5.459e-03 9.560e-03 2.430e+02
## t value Pr(>|t|)
## (Intercept) -2.337 0.0204 *
## beforeafteron 0.653 0.5144
## conditioncontrol 1.840 0.0670 .
## conditionmeaningfully 1.804 0.0725 .
## conditionmeaninglessly 1.529 0.1276
## conditionrepeat 2.310 0.0217 *
## age.c -0.162 0.8718
## beforeafteron:conditioncontrol -0.630 0.5291
## beforeafteron:conditionmeaningfully -1.523 0.1292
## beforeafteron:conditionmeaninglessly -1.082 0.2804
## beforeafteron:conditionrepeat -0.534 0.5941
## beforeafteron:age.c 0.554 0.5803
## conditioncontrol:age.c 0.029 0.9771
## conditionmeaningfully:age.c -0.102 0.9185
## conditionmeaninglessly:age.c -0.607 0.5441
## conditionrepeat:age.c 0.041 0.9676
## beforeafteron:conditioncontrol:age.c -0.479 0.6320
## beforeafteron:conditionmeaningfully:age.c 0.443 0.6583
## beforeafteron:conditionmeaninglessly:age.c 0.152 0.8795
## beforeafteron:conditionrepeat:age.c -0.571 0.5685
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Correlation matrix not shown by default, as p = 20 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')# calculate the contribution of age
## Get interaction coefficient between age and beforeafter
bXa.coeff <- fixef(Model)[12]
## Random effect in model for beforeafter - squared to compute variance
residvalvar <- 0.002668
## Calculate the implied total random effect variance - interaction squared multiplied by variance in age + residual variance.
imptotalvalvar <- bXa.coeff^2*sd(RESPdata$age.c)^2 + residvalvar
## Variance in mu difference explained by age
V_RESP <- 1 - (residvalvar/imptotalvalvar)
V_RESP ## beforeafteron:age.c
## 0.3597392# plot the data
## computing random effects
ageranef_original <- as.data.frame(ranef(Model)) %>%
dplyr::select("grp","term","condval")
library(reshape2)
ageranef <- dcast(ageranef_original,
grp~term,
timevar = c("condval")) ## Warning in dcast(ageranef_original, grp ~ term, timevar = c("condval")): The
## dcast generic in data.table has been passed a data.frame and will attempt to
## redirect to the relevant reshape2 method; please note that reshape2 is
## superseded and is no longer actively developed, and this redirection is now
## deprecated. Please do this redirection yourself like
## reshape2::dcast(ageranef_original). In the next version, this warning will
## become an error.## Using 'condval' as value column. Use 'value.var' to overrideageranef <- ageranef %>% rename("intercept_age"="(Intercept)","slope_age"="beforeafteron","sub"="grp")
## Create dataset with one line per person with mileage score
ageranef <- merge(ageranef, RESPdata, by = 'sub')
ageranef <- ageranef %>%
dplyr::group_by(sub) %>%
dplyr::slice(1) %>%
dplyr::select(sub, age.c, intercept_age, slope_age)
## Person-specific implied total n-back effects for model accounting for age
ranef.age.pred.tot <- fixef(Model)[2] + # fixed effect for n-back
fixef(Model)[12] * ageranef$age.c + # fixed effect for n-back X age interaction # scale of age
ageranef$slope_age # estimate random effect of slopes
## relationship between the effect of N-back and age
age.nback.effect.sge_breathingrate <- ggplot(ageranef, aes(age.c, ranef.age.pred.tot)) +
geom_vline(xintercept = 0, size = .5, color = "black", linetype="solid") +
geom_hline(yintercept = mean(ranef.age.pred.tot), size = .5, color = "black", linetype="solid") +
geom_jitter(height = 0, width = 0, size = 4,
shape = 21, colour = "black", fill = viridis(5)[3], alpha = .95, stroke = 1) +
stat_smooth(method = 'lm', ci = T, color = viridis(5)[3], size = 1) +
xlab("Age (mean centered)") +
ylab(expression("Implied total heterogeneity of "~ breathing~rate)) +
#ylim(-.15, .1) +
theme_bw() +
theme(legend.position = "bottom", legend.direction = "horizontal", axis.title.x = element_text(size = 11), axis.text.x = element_text(size = 15), axis.title.y = element_text(size = 11), axis.text.y = element_text(size = 15), title = element_text(size = 18), legend.title = element_text(size = 12), legend.text = element_text(size = 12), legend.key = element_blank(), legend.key.width = unit(0.3, 'cm'), legend.key.size = unit(0.1, 'cm'), plot.title = element_text(size = 7.5, face = "bold"), strip.text = element_text(face = "bold", size = 10))## Warning in stat_smooth(method = "lm", ci = T, color = viridis(5)[3], size = 1):
## Ignoring unknown parameters: `ci`age.nback.effect.sge_breathingrate## `geom_smooth()` using formula = 'y ~ x'
PD
# scale the data
PDdata$age.c <- scale(PDdata$age, center = T, scale = F)
# fit model
Model <- lmer(data = PDdata, PD~beforeafter*condition*age.c+(1+beforeafter|sub))## boundary (singular) fit: see help('isSingular')summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: PD ~ beforeafter * condition * age.c + (1 + beforeafter | sub)
## Data: PDdata
##
## REML criterion at convergence: -552.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.3610 -0.5808 -0.0081 0.6074 3.2957
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## sub (Intercept) 0.0006927 0.02632
## beforeafteron 0.0003204 0.01790 1.00
## Residual 0.0043103 0.06565
## Number of obs: 290, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df
## (Intercept) 9.990e-01 1.313e-02 1.780e+02
## beforeafteron 1.253e-01 1.756e-02 2.264e+02
## conditioncontrol 2.310e-03 1.724e-02 2.430e+02
## conditionmeaningfully -1.559e-03 1.724e-02 2.430e+02
## conditionmeaninglessly 1.131e-02 1.724e-02 2.430e+02
## conditionrepeat -8.319e-03 1.724e-02 2.430e+02
## age.c -3.163e-04 1.284e-03 1.781e+02
## beforeafteron:conditioncontrol -1.300e-01 2.438e-02 2.430e+02
## beforeafteron:conditionmeaningfully -1.137e-01 2.438e-02 2.430e+02
## beforeafteron:conditionmeaninglessly -1.362e-01 2.438e-02 2.430e+02
## beforeafteron:conditionrepeat -3.783e-02 2.438e-02 2.430e+02
## beforeafteron:age.c 1.953e-03 1.717e-03 2.264e+02
## conditioncontrol:age.c 8.821e-04 1.686e-03 2.430e+02
## conditionmeaningfully:age.c 9.116e-04 1.685e-03 2.430e+02
## conditionmeaninglessly:age.c 3.650e-04 1.686e-03 2.430e+02
## conditionrepeat:age.c -2.636e-04 1.686e-03 2.430e+02
## beforeafteron:conditioncontrol:age.c -1.161e-03 2.384e-03 2.430e+02
## beforeafteron:conditionmeaningfully:age.c -2.319e-03 2.383e-03 2.430e+02
## beforeafteron:conditionmeaninglessly:age.c -7.198e-04 2.384e-03 2.430e+02
## beforeafteron:conditionrepeat:age.c -1.561e-03 2.384e-03 2.430e+02
## t value Pr(>|t|)
## (Intercept) 76.059 < 2e-16 ***
## beforeafteron 7.137 1.28e-11 ***
## conditioncontrol 0.134 0.894
## conditionmeaningfully -0.090 0.928
## conditionmeaninglessly 0.656 0.512
## conditionrepeat -0.483 0.630
## age.c -0.246 0.806
## beforeafteron:conditioncontrol -5.332 2.22e-07 ***
## beforeafteron:conditionmeaningfully -4.664 5.12e-06 ***
## beforeafteron:conditionmeaninglessly -5.588 6.16e-08 ***
## beforeafteron:conditionrepeat -1.551 0.122
## beforeafteron:age.c 1.137 0.257
## conditioncontrol:age.c 0.523 0.601
## conditionmeaningfully:age.c 0.541 0.589
## conditionmeaninglessly:age.c 0.217 0.829
## conditionrepeat:age.c -0.156 0.876
## beforeafteron:conditioncontrol:age.c -0.487 0.627
## beforeafteron:conditionmeaningfully:age.c -0.973 0.331
## beforeafteron:conditionmeaninglessly:age.c -0.302 0.763
## beforeafteron:conditionrepeat:age.c -0.655 0.513
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Correlation matrix not shown by default, as p = 20 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')# calculate the contribution of age
# Get interaction coefficient between age and beforeafter
bXa.coeff <- fixef(Model)[12]
# Random effect in model for beforeafter - squared to compute variance
residvalvar <- 0.0003204
# Calculate the implied total random effect variance - interaction squared multiplied by variance in age + residual variance.
imptotalvalvar <- bXa.coeff^2*sd(PDdata$age.c)^2 + residvalvar
# Variance in mu difference explained by age
V_PD <- 1 - (residvalvar/imptotalvalvar)
V_PD## beforeafteron:age.c
## 0.5555394# plot the data
## computing random effects
ageranef_original <- as.data.frame(ranef(Model)) %>%
dplyr::select("grp","term","condval")
library(reshape2)
ageranef <- dcast(ageranef_original,
grp~term,
timevar = c("condval")) ## Warning in dcast(ageranef_original, grp ~ term, timevar = c("condval")): The
## dcast generic in data.table has been passed a data.frame and will attempt to
## redirect to the relevant reshape2 method; please note that reshape2 is
## superseded and is no longer actively developed, and this redirection is now
## deprecated. Please do this redirection yourself like
## reshape2::dcast(ageranef_original). In the next version, this warning will
## become an error.## Using 'condval' as value column. Use 'value.var' to overrideageranef <- ageranef %>% rename("intercept_age"="(Intercept)","slope_age"="beforeafteron","sub"="grp")
## Create dataset with one line per person with mileage score
ageranef <- merge(ageranef, PDdata, by = 'sub')
ageranef <- ageranef %>%
dplyr::group_by(sub) %>%
dplyr::slice(1) %>%
dplyr::select(sub, age.c, intercept_age, slope_age)
## Person-specific implied total n-back effects for model accounting for age
ranef.age.pred.tot <- fixef(Model)[2] + # fixed effect for n-back
fixef(Model)[12] * ageranef$age.c + # fixed effect for n-back X age interaction # scale of age
ageranef$slope_age # estimate random effect of slopes
## relationship between the effect of N-back and age
age.nback.effect.sge_pupildiameter <- ggplot(ageranef, aes(age.c, ranef.age.pred.tot)) +
geom_vline(xintercept = 0, size = .5, color = "black", linetype="solid") +
geom_hline(yintercept = mean(ranef.age.pred.tot), size = .5, color = "black", linetype="solid") +
geom_jitter(height = 0, width = 0, size = 4,
shape = 21, colour = "black", fill = viridis(5)[3], alpha = .95, stroke = 1) +
stat_smooth(method = 'lm', ci = T, color = viridis(5)[3], size = 1) +
xlab("Age (mean centered)") +
ylab(expression("Implied total heterogeneity of "~ pupil~diameter)) +
#ylim(-.15, .1) +
theme_bw() +
theme(legend.position = "bottom", legend.direction = "horizontal", axis.title.x = element_text(size = 11), axis.text.x = element_text(size = 15), axis.title.y = element_text(size = 11), axis.text.y = element_text(size = 15), title = element_text(size = 18), legend.title = element_text(size = 12), legend.text = element_text(size = 12), legend.key = element_blank(), legend.key.width = unit(0.3, 'cm'), legend.key.size = unit(0.1, 'cm'), plot.title = element_text(size = 7.5, face = "bold"), strip.text = element_text(face = "bold", size = 10))## Warning in stat_smooth(method = "lm", ci = T, color = viridis(5)[3], size = 1):
## Ignoring unknown parameters: `ci`age.nback.effect.sge_pupildiameter## `geom_smooth()` using formula = 'y ~ x'
### PVRC
# scale the data
PVRCdata$age.c <- scale(PVRCdata$age, center = T, scale = F)
# fit model
Model <- lmer(data = PVRCdata, PVRC~beforeafter*condition*age.c+(1+beforeafter|sub))## boundary (singular) fit: see help('isSingular')summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: PVRC ~ beforeafter * condition * age.c + (1 + beforeafter | sub)
## Data: PVRCdata
##
## REML criterion at convergence: 278.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.3607 -0.5027 -0.0113 0.4194 4.8110
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## sub (Intercept) 0.0104590 0.10227
## beforeafteron 0.0002471 0.01572 -1.00
## Residual 0.1010921 0.31795
## Number of obs: 290, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df
## (Intercept) 8.433e-01 6.202e-02 2.059e+02
## beforeafteron 1.533e-01 8.355e-02 2.427e+02
## conditioncontrol 1.836e-01 8.350e-02 2.430e+02
## conditionmeaningfully 1.030e-01 8.350e-02 2.430e+02
## conditionmeaninglessly 6.242e-02 8.350e-02 2.430e+02
## conditionrepeat 4.500e-02 8.350e-02 2.430e+02
## age.c -4.689e-04 6.065e-03 2.059e+02
## beforeafteron:conditioncontrol -3.022e-01 1.181e-01 2.430e+02
## beforeafteron:conditionmeaningfully -9.015e-02 1.181e-01 2.430e+02
## beforeafteron:conditionmeaninglessly -8.301e-02 1.181e-01 2.430e+02
## beforeafteron:conditionrepeat 8.212e-02 1.181e-01 2.430e+02
## beforeafteron:age.c 3.387e-03 8.170e-03 2.427e+02
## conditioncontrol:age.c -3.443e-03 8.165e-03 2.430e+02
## conditionmeaningfully:age.c -3.218e-03 8.161e-03 2.430e+02
## conditionmeaninglessly:age.c -2.004e-04 8.165e-03 2.430e+02
## conditionrepeat:age.c -1.424e-03 8.165e-03 2.430e+02
## beforeafteron:conditioncontrol:age.c 4.210e-03 1.155e-02 2.430e+02
## beforeafteron:conditionmeaningfully:age.c -2.011e-03 1.154e-02 2.430e+02
## beforeafteron:conditionmeaninglessly:age.c 2.221e-03 1.155e-02 2.430e+02
## beforeafteron:conditionrepeat:age.c -9.363e-03 1.155e-02 2.430e+02
## t value Pr(>|t|)
## (Intercept) 13.597 <2e-16 ***
## beforeafteron 1.835 0.0677 .
## conditioncontrol 2.199 0.0288 *
## conditionmeaningfully 1.234 0.2184
## conditionmeaninglessly 0.748 0.4554
## conditionrepeat 0.539 0.5904
## age.c -0.077 0.9385
## beforeafteron:conditioncontrol -2.559 0.0111 *
## beforeafteron:conditionmeaningfully -0.763 0.4459
## beforeafteron:conditionmeaninglessly -0.703 0.4827
## beforeafteron:conditionrepeat 0.695 0.4875
## beforeafteron:age.c 0.415 0.6789
## conditioncontrol:age.c -0.422 0.6736
## conditionmeaningfully:age.c -0.394 0.6937
## conditionmeaninglessly:age.c -0.025 0.9804
## conditionrepeat:age.c -0.174 0.8617
## beforeafteron:conditioncontrol:age.c 0.365 0.7157
## beforeafteron:conditionmeaningfully:age.c -0.174 0.8618
## beforeafteron:conditionmeaninglessly:age.c 0.192 0.8476
## beforeafteron:conditionrepeat:age.c -0.811 0.4182
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Correlation matrix not shown by default, as p = 20 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')# calculate the contribution of age
# Get interaction coefficient between age and beforeafter
bXa.coeff <- fixef(Model)[12]
# Random effect in model for beforeafter - squared to compute variance
residvalvar <- 0.0002471
# Calculate the implied total random effect variance - interaction squared multiplied by variance in age + residual variance.
imptotalvalvar <- bXa.coeff^2*sd(PVRCdata$age.c)^2 + residvalvar
# Variance in mu difference explained by age
V_PVRC <- 1 - (residvalvar/imptotalvalvar)
V_PVRC ## beforeafteron:age.c
## 0.8297245# plot the data
## computing random effects
ageranef_original <- as.data.frame(ranef(Model)) %>%
dplyr::select("grp","term","condval")
library(reshape2)
ageranef <- dcast(ageranef_original,
grp~term,
timevar = c("condval")) ## Warning in dcast(ageranef_original, grp ~ term, timevar = c("condval")): The
## dcast generic in data.table has been passed a data.frame and will attempt to
## redirect to the relevant reshape2 method; please note that reshape2 is
## superseded and is no longer actively developed, and this redirection is now
## deprecated. Please do this redirection yourself like
## reshape2::dcast(ageranef_original). In the next version, this warning will
## become an error.## Using 'condval' as value column. Use 'value.var' to overrideageranef <- ageranef %>% rename("intercept_age"="(Intercept)","slope_age"="beforeafteron","sub"="grp")
## Create dataset with one line per person with mileage score
ageranef <- merge(ageranef, PVRCdata, by = 'sub')
ageranef <- ageranef %>%
dplyr::group_by(sub) %>%
dplyr::slice(1) %>%
dplyr::select(sub, age.c, intercept_age, slope_age)
## Person-specific implied total n-back effects for model accounting for age
ranef.age.pred.tot <- fixef(Model)[2] + # fixed effect for n-back
fixef(Model)[12] * ageranef$age.c + # fixed effect for n-back X age interaction # scale of age
ageranef$slope_age # estimate random effect of slopes
## relationship between the effect of N-back and age
age.nback.effect.sge_PVRC <- ggplot(ageranef, aes(age.c, ranef.age.pred.tot)) +
geom_vline(xintercept = 0, size = .5, color = "black", linetype="solid") +
geom_hline(yintercept = mean(ranef.age.pred.tot), size = .5, color = "black", linetype="solid") +
geom_jitter(height = 0, width = 0, size = 4,
shape = 21, colour = "black", fill = viridis(5)[3], alpha = .95, stroke = 1) +
stat_smooth(method = 'lm', ci = T, color = viridis(5)[3], size = 1) +
xlab("Age (mean centered)") +
ylab(expression("Implied total heterogeneity of "~ PVRC)) +
#ylim(-.15, .1) +
theme_bw() +
theme(legend.position = "bottom", legend.direction = "horizontal", axis.title.x = element_text(size = 11), axis.text.x = element_text(size = 15), axis.title.y = element_text(size = 11), axis.text.y = element_text(size = 15), title = element_text(size = 18), legend.title = element_text(size = 12), legend.text = element_text(size = 12), legend.key = element_blank(), legend.key.width = unit(0.3, 'cm'), legend.key.size = unit(0.1, 'cm'), plot.title = element_text(size = 7.5, face = "bold"), strip.text = element_text(face = "bold", size = 10))## Warning in stat_smooth(method = "lm", ci = T, color = viridis(5)[3], size = 1):
## Ignoring unknown parameters: `ci`age.nback.effect.sge_PVRC## `geom_smooth()` using formula = 'y ~ x'
Blink rate (BL)
# scale the data
BLdata$age.c <- scale(BLdata$age, center = T, scale = F)
# fit model
Model <- lmer(data = BLdata, BL~beforeafter*condition*age.c+(1+beforeafter|sub))## boundary (singular) fit: see help('isSingular')summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: BL ~ beforeafter * condition * age.c + (1 + beforeafter | sub)
## Data: BLdata
##
## REML criterion at convergence: 560.7
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -5.4895 -0.5381 -0.0099 0.4460 3.1664
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## sub (Intercept) 6.403e-02 0.253044
## beforeafteron 2.304e-06 0.001518 -1.00
## Residual 2.716e-01 0.521125
## Number of obs: 290, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error df
## (Intercept) 0.198663 0.107576 155.838350
## beforeafteron 0.130084 0.136854 242.997068
## conditioncontrol -0.086656 0.136854 242.998986
## conditionmeaningfully 0.139762 0.136855 242.999914
## conditionmeaninglessly -0.006610 0.136854 242.998986
## conditionrepeat -0.030366 0.136854 242.998986
## age.c -0.006783 0.010519 155.866888
## beforeafteron:conditioncontrol -0.131590 0.193541 242.998986
## beforeafteron:conditionmeaningfully -0.109088 0.193541 242.998986
## beforeafteron:conditionmeaninglessly -0.135853 0.193541 242.998986
## beforeafteron:conditionrepeat -0.007070 0.193541 242.998986
## beforeafteron:age.c 0.003667 0.013383 242.997066
## conditioncontrol:age.c 0.004862 0.013383 242.998986
## conditionmeaningfully:age.c 0.007849 0.013375 243.001420
## conditionmeaninglessly:age.c 0.015347 0.013383 242.998986
## conditionrepeat:age.c 0.018676 0.013383 242.998986
## beforeafteron:conditioncontrol:age.c -0.009054 0.018926 242.998986
## beforeafteron:conditionmeaningfully:age.c -0.005097 0.018915 242.998987
## beforeafteron:conditionmeaninglessly:age.c -0.012895 0.018926 242.998986
## beforeafteron:conditionrepeat:age.c -0.007414 0.018926 242.998986
## t value Pr(>|t|)
## (Intercept) 1.847 0.0667 .
## beforeafteron 0.951 0.3428
## conditioncontrol -0.633 0.5272
## conditionmeaningfully 1.021 0.3082
## conditionmeaninglessly -0.048 0.9615
## conditionrepeat -0.222 0.8246
## age.c -0.645 0.5200
## beforeafteron:conditioncontrol -0.680 0.4972
## beforeafteron:conditionmeaningfully -0.564 0.5735
## beforeafteron:conditionmeaninglessly -0.702 0.4834
## beforeafteron:conditionrepeat -0.037 0.9709
## beforeafteron:age.c 0.274 0.7843
## conditioncontrol:age.c 0.363 0.7167
## conditionmeaningfully:age.c 0.587 0.5579
## conditionmeaninglessly:age.c 1.147 0.2526
## conditionrepeat:age.c 1.396 0.1641
## beforeafteron:conditioncontrol:age.c -0.478 0.6328
## beforeafteron:conditionmeaningfully:age.c -0.269 0.7878
## beforeafteron:conditionmeaninglessly:age.c -0.681 0.4963
## beforeafteron:conditionrepeat:age.c -0.392 0.6956
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Correlation matrix not shown by default, as p = 20 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need it## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')# calculate the contribution of age
# Get interaction coefficient between age and beforeafter
bXa.coeff <- fixef(Model)[12]
# Random effect in model for beforeafter - squared to compute variance
residvalvar <- 2.304e-06
# Calculate the implied total random effect variance - interaction squared multiplied by variance in age + residual variance.
imptotalvalvar <- bXa.coeff^2*sd(BLdata$age.c)^2 + residvalvar
# Variance in mu difference explained by age
V_BL <- 1 - (residvalvar/imptotalvalvar)
V_BL## beforeafteron:age.c
## 0.9983708# plot the data
## computing random effects
ageranef_original <- as.data.frame(ranef(Model)) %>%
dplyr::select("grp","term","condval")
library(reshape2)
ageranef <- dcast(ageranef_original,
grp~term,
timevar = c("condval"))## Warning in dcast(ageranef_original, grp ~ term, timevar = c("condval")): The
## dcast generic in data.table has been passed a data.frame and will attempt to
## redirect to the relevant reshape2 method; please note that reshape2 is
## superseded and is no longer actively developed, and this redirection is now
## deprecated. Please do this redirection yourself like
## reshape2::dcast(ageranef_original). In the next version, this warning will
## become an error.## Using 'condval' as value column. Use 'value.var' to overrideageranef <- ageranef %>% rename("intercept_age"="(Intercept)","slope_age"="beforeafteron","sub"="grp")
## Create dataset with one line per person with mileage score
ageranef <- merge(ageranef, BLdata, by = 'sub')
ageranef <- ageranef %>%
dplyr::group_by(sub) %>%
dplyr::slice(1) %>%
dplyr::select(sub, age.c, intercept_age, slope_age)
## Person-specific implied total n-back effects for model accounting for age
ranef.age.pred.tot <- fixef(Model)[2] + # fixed effect for n-back
fixef(Model)[12] * ageranef$age.c + # fixed effect for n-back X age interaction # scale of age
ageranef$slope_age # estimate random effect of slopes
## relationship between the effect of N-back and age
age.nback.effect.sge_blinkrate <- ggplot(ageranef, aes(age.c, ranef.age.pred.tot)) +
geom_vline(xintercept = 0, size = .5, color = "black", linetype="solid") +
geom_hline(yintercept = mean(ranef.age.pred.tot), size = .5, color = "black", linetype="solid") +
geom_jitter(height = 0, width = 0, size = 4,
shape = 21, colour = "black", fill = viridis(5)[3], alpha = .95, stroke = 1) +
stat_smooth(method = 'lm', ci = T, color = viridis(5)[3], size = 1) +
xlab("Age (mean centered)") +
ylab(expression("Implied total heterogeneity of "~ blink~rate)) +
#ylim(-.15, .1) +
theme_bw() +
theme(legend.position = "bottom", legend.direction = "horizontal", axis.title.x = element_text(size = 11), axis.text.x = element_text(size = 15), axis.title.y = element_text(size = 11), axis.text.y = element_text(size = 15), title = element_text(size = 18), legend.title = element_text(size = 12), legend.text = element_text(size = 12), legend.key = element_blank(), legend.key.width = unit(0.3, 'cm'), legend.key.size = unit(0.1, 'cm'), plot.title = element_text(size = 7.5, face = "bold"), strip.text = element_text(face = "bold", size = 10))## Warning in stat_smooth(method = "lm", ci = T, color = viridis(5)[3], size = 1):
## Ignoring unknown parameters: `ci`age.nback.effect.sge_blinkrate## `geom_smooth()` using formula = 'y ~ x'
3.6 Subjective ratings
3.6.1 Workload
# read data
MWdata <- read_excel("/Users/betty/Desktop/Workload(processed).xlsx", sheet=1)
# fit model
Model <- lmer(data = MWdata, meanscore ~ agegroup2*condition*dimension + (1|sub))
summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: meanscore ~ agegroup2 * condition * dimension + (1 | sub)
## Data: MWdata
##
## REML criterion at convergence: 7615.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.1452 -0.6837 -0.0384 0.6863 2.3401
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 103.8 10.19
## Residual 545.9 23.36
## Number of obs: 870, groups: sub, 29
##
## Fixed effects:
## Estimate Std. Error
## (Intercept) 52.8571 6.8121
## agegroup2Y -6.8571 9.4718
## conditioncontrol 0.7143 8.8310
## conditionmeaningfully -8.5714 8.8310
## conditionmeaninglessly 7.8571 8.8310
## conditionrepeat -1.7033 8.8310
## dimensionfrustration -19.2857 8.8310
## dimensionmental -0.7143 8.8310
## dimensionperformance -16.0714 8.8310
## dimensionphysical -10.0000 8.8310
## dimensiontime -5.7143 8.8310
## agegroup2Y:conditioncontrol 3.6190 12.2790
## agegroup2Y:conditionmeaningfully 10.9048 12.2790
## agegroup2Y:conditionmeaninglessly -5.1905 12.2790
## agegroup2Y:conditionrepeat -11.2967 12.2790
## agegroup2Y:dimensionfrustration -8.3810 12.2790
## agegroup2Y:dimensionmental 13.3810 12.2790
## agegroup2Y:dimensionperformance 5.4048 12.2790
## agegroup2Y:dimensionphysical 1.0000 12.2790
## agegroup2Y:dimensiontime -2.2857 12.2790
## conditioncontrol:dimensionfrustration 2.8571 12.4889
## conditionmeaningfully:dimensionfrustration 3.9286 12.4889
## conditionmeaninglessly:dimensionfrustration -4.6429 12.4889
## conditionrepeat:dimensionfrustration -0.7143 12.4889
## conditioncontrol:dimensionmental -0.7143 12.4889
## conditionmeaningfully:dimensionmental 17.1429 12.4889
## conditionmeaninglessly:dimensionmental -13.2143 12.4889
## conditionrepeat:dimensionmental -3.5165 12.4889
## conditioncontrol:dimensionperformance 6.4286 12.4889
## conditionmeaningfully:dimensionperformance 9.6429 12.4889
## conditionmeaninglessly:dimensionperformance -1.4286 12.4889
## conditionrepeat:dimensionperformance 3.7637 12.4889
## conditioncontrol:dimensionphysical 6.0714 12.4889
## conditionmeaningfully:dimensionphysical 15.0000 12.4889
## conditionmeaninglessly:dimensionphysical -6.7857 12.4889
## conditionrepeat:dimensionphysical -3.0769 12.4889
## conditioncontrol:dimensiontime -8.9286 12.4889
## conditionmeaningfully:dimensiontime -2.5000 12.4889
## conditionmeaninglessly:dimensiontime -6.7857 12.4889
## conditionrepeat:dimensiontime -0.4396 12.4889
## agegroup2Y:conditioncontrol:dimensionfrustration 15.1429 17.3651
## agegroup2Y:conditionmeaningfully:dimensionfrustration 20.4048 17.3651
## agegroup2Y:conditionmeaninglessly:dimensionfrustration 30.3095 17.3651
## agegroup2Y:conditionrepeat:dimensionfrustration 19.0476 17.3651
## agegroup2Y:conditioncontrol:dimensionmental -25.6190 17.3651
## agegroup2Y:conditionmeaningfully:dimensionmental -44.8095 17.3651
## agegroup2Y:conditionmeaninglessly:dimensionmental -18.7857 17.3651
## agegroup2Y:conditionrepeat:dimensionmental 4.5165 17.3651
## agegroup2Y:conditioncontrol:dimensionperformance 2.9048 17.3651
## agegroup2Y:conditionmeaningfully:dimensionperformance 5.6905 17.3651
## agegroup2Y:conditionmeaninglessly:dimensionperformance 12.7619 17.3651
## agegroup2Y:conditionrepeat:dimensionperformance 4.9030 17.3651
## agegroup2Y:conditioncontrol:dimensionphysical -1.0714 17.3651
## agegroup2Y:conditionmeaningfully:dimensionphysical -21.3333 17.3651
## agegroup2Y:conditionmeaninglessly:dimensionphysical 5.7857 17.3651
## agegroup2Y:conditionrepeat:dimensionphysical 15.7436 17.3651
## agegroup2Y:conditioncontrol:dimensiontime 6.5952 17.3651
## agegroup2Y:conditionmeaningfully:dimensiontime 6.8333 17.3651
## agegroup2Y:conditionmeaninglessly:dimensiontime 10.4524 17.3651
## agegroup2Y:conditionrepeat:dimensiontime 3.1062 17.3651
## df t value
## (Intercept) 465.5954 7.759
## agegroup2Y 465.5954 -0.724
## conditioncontrol 783.0000 0.081
## conditionmeaningfully 783.0000 -0.971
## conditionmeaninglessly 783.0000 0.890
## conditionrepeat 783.0000 -0.193
## dimensionfrustration 783.0000 -2.184
## dimensionmental 783.0000 -0.081
## dimensionperformance 783.0000 -1.820
## dimensionphysical 783.0000 -1.132
## dimensiontime 783.0000 -0.647
## agegroup2Y:conditioncontrol 783.0000 0.295
## agegroup2Y:conditionmeaningfully 783.0000 0.888
## agegroup2Y:conditionmeaninglessly 783.0000 -0.423
## agegroup2Y:conditionrepeat 783.0000 -0.920
## agegroup2Y:dimensionfrustration 783.0000 -0.683
## agegroup2Y:dimensionmental 783.0000 1.090
## agegroup2Y:dimensionperformance 783.0000 0.440
## agegroup2Y:dimensionphysical 783.0000 0.081
## agegroup2Y:dimensiontime 783.0000 -0.186
## conditioncontrol:dimensionfrustration 783.0000 0.229
## conditionmeaningfully:dimensionfrustration 783.0000 0.315
## conditionmeaninglessly:dimensionfrustration 783.0000 -0.372
## conditionrepeat:dimensionfrustration 783.0000 -0.057
## conditioncontrol:dimensionmental 783.0000 -0.057
## conditionmeaningfully:dimensionmental 783.0000 1.373
## conditionmeaninglessly:dimensionmental 783.0000 -1.058
## conditionrepeat:dimensionmental 783.0000 -0.282
## conditioncontrol:dimensionperformance 783.0000 0.515
## conditionmeaningfully:dimensionperformance 783.0000 0.772
## conditionmeaninglessly:dimensionperformance 783.0000 -0.114
## conditionrepeat:dimensionperformance 783.0000 0.301
## conditioncontrol:dimensionphysical 783.0000 0.486
## conditionmeaningfully:dimensionphysical 783.0000 1.201
## conditionmeaninglessly:dimensionphysical 783.0000 -0.543
## conditionrepeat:dimensionphysical 783.0000 -0.246
## conditioncontrol:dimensiontime 783.0000 -0.715
## conditionmeaningfully:dimensiontime 783.0000 -0.200
## conditionmeaninglessly:dimensiontime 783.0000 -0.543
## conditionrepeat:dimensiontime 783.0000 -0.035
## agegroup2Y:conditioncontrol:dimensionfrustration 783.0000 0.872
## agegroup2Y:conditionmeaningfully:dimensionfrustration 783.0000 1.175
## agegroup2Y:conditionmeaninglessly:dimensionfrustration 783.0000 1.745
## agegroup2Y:conditionrepeat:dimensionfrustration 783.0000 1.097
## agegroup2Y:conditioncontrol:dimensionmental 783.0000 -1.475
## agegroup2Y:conditionmeaningfully:dimensionmental 783.0000 -2.580
## agegroup2Y:conditionmeaninglessly:dimensionmental 783.0000 -1.082
## agegroup2Y:conditionrepeat:dimensionmental 783.0000 0.260
## agegroup2Y:conditioncontrol:dimensionperformance 783.0000 0.167
## agegroup2Y:conditionmeaningfully:dimensionperformance 783.0000 0.328
## agegroup2Y:conditionmeaninglessly:dimensionperformance 783.0000 0.735
## agegroup2Y:conditionrepeat:dimensionperformance 783.0000 0.282
## agegroup2Y:conditioncontrol:dimensionphysical 783.0000 -0.062
## agegroup2Y:conditionmeaningfully:dimensionphysical 783.0000 -1.229
## agegroup2Y:conditionmeaninglessly:dimensionphysical 783.0000 0.333
## agegroup2Y:conditionrepeat:dimensionphysical 783.0000 0.907
## agegroup2Y:conditioncontrol:dimensiontime 783.0000 0.380
## agegroup2Y:conditionmeaningfully:dimensiontime 783.0000 0.394
## agegroup2Y:conditionmeaninglessly:dimensiontime 783.0000 0.602
## agegroup2Y:conditionrepeat:dimensiontime 783.0000 0.179
## Pr(>|t|)
## (Intercept) 5.45e-14 ***
## agegroup2Y 0.4695
## conditioncontrol 0.9356
## conditionmeaningfully 0.3320
## conditionmeaninglessly 0.3739
## conditionrepeat 0.8471
## dimensionfrustration 0.0293 *
## dimensionmental 0.9356
## dimensionperformance 0.0692 .
## dimensionphysical 0.2578
## dimensiontime 0.5178
## agegroup2Y:conditioncontrol 0.7683
## agegroup2Y:conditionmeaningfully 0.3748
## agegroup2Y:conditionmeaninglessly 0.6726
## agegroup2Y:conditionrepeat 0.3579
## agegroup2Y:dimensionfrustration 0.4951
## agegroup2Y:dimensionmental 0.2762
## agegroup2Y:dimensionperformance 0.6599
## agegroup2Y:dimensionphysical 0.9351
## agegroup2Y:dimensiontime 0.8524
## conditioncontrol:dimensionfrustration 0.8191
## conditionmeaningfully:dimensionfrustration 0.7532
## conditionmeaninglessly:dimensionfrustration 0.7102
## conditionrepeat:dimensionfrustration 0.9544
## conditioncontrol:dimensionmental 0.9544
## conditionmeaningfully:dimensionmental 0.1703
## conditionmeaninglessly:dimensionmental 0.2903
## conditionrepeat:dimensionmental 0.7783
## conditioncontrol:dimensionperformance 0.6069
## conditionmeaningfully:dimensionperformance 0.4403
## conditionmeaninglessly:dimensionperformance 0.9090
## conditionrepeat:dimensionperformance 0.7632
## conditioncontrol:dimensionphysical 0.6270
## conditionmeaningfully:dimensionphysical 0.2301
## conditionmeaninglessly:dimensionphysical 0.5871
## conditionrepeat:dimensionphysical 0.8055
## conditioncontrol:dimensiontime 0.4749
## conditionmeaningfully:dimensiontime 0.8414
## conditionmeaninglessly:dimensiontime 0.5871
## conditionrepeat:dimensiontime 0.9719
## agegroup2Y:conditioncontrol:dimensionfrustration 0.3835
## agegroup2Y:conditionmeaningfully:dimensionfrustration 0.2403
## agegroup2Y:conditionmeaninglessly:dimensionfrustration 0.0813 .
## agegroup2Y:conditionrepeat:dimensionfrustration 0.2730
## agegroup2Y:conditioncontrol:dimensionmental 0.1405
## agegroup2Y:conditionmeaningfully:dimensionmental 0.0100 *
## agegroup2Y:conditionmeaninglessly:dimensionmental 0.2797
## agegroup2Y:conditionrepeat:dimensionmental 0.7949
## agegroup2Y:conditioncontrol:dimensionperformance 0.8672
## agegroup2Y:conditionmeaningfully:dimensionperformance 0.7432
## agegroup2Y:conditionmeaninglessly:dimensionperformance 0.4626
## agegroup2Y:conditionrepeat:dimensionperformance 0.7778
## agegroup2Y:conditioncontrol:dimensionphysical 0.9508
## agegroup2Y:conditionmeaningfully:dimensionphysical 0.2196
## agegroup2Y:conditionmeaninglessly:dimensionphysical 0.7391
## agegroup2Y:conditionrepeat:dimensionphysical 0.3649
## agegroup2Y:conditioncontrol:dimensiontime 0.7042
## agegroup2Y:conditionmeaningfully:dimensiontime 0.6941
## agegroup2Y:conditionmeaninglessly:dimensiontime 0.5474
## agegroup2Y:conditionrepeat:dimensiontime 0.8581
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Correlation matrix not shown by default, as p = 60 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need itanova(Model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## agegroup2 512.8 512.8 1 27 0.9393 0.34106
## condition 6341.8 1585.4 4 783 2.9042 0.02106 *
## dimension 18533.7 3706.7 5 783 6.7901 3.283e-06 ***
## agegroup2:condition 1929.1 482.3 4 783 0.8834 0.47329
## agegroup2:dimension 5297.1 1059.4 5 783 1.9406 0.08540 .
## condition:dimension 14045.6 702.3 20 783 1.2864 0.17927
## agegroup2:condition:dimension 16217.7 810.9 20 783 1.4854 0.07852 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1# the post hoc analysis for the main effect of countermeasure type
pwc.workload.condition <- MWdata %>% pairwise_t_test(meanscore ~ condition, p.adjust.method = "bonferroni", paired = TRUE)
pwc.workload.condition ## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 meanscore answer contr… 174 174 -1.48 173 0.14 1 ns
## 2 meanscore answer meani… 174 174 -0.601 173 0.549 1 ns
## 3 meanscore answer meani… 174 174 -1.12 173 0.266 1 ns
## 4 meanscore answer repeat 174 174 1.99 173 0.048 0.477 ns
## 5 meanscore control meani… 174 174 0.912 173 0.363 1 ns
## 6 meanscore control meani… 174 174 0.0695 173 0.945 1 ns
## 7 meanscore control repeat 174 174 3.26 173 0.001 0.013 *
## 8 meanscore meanin… meani… 174 174 -0.737 173 0.462 1 ns
## 9 meanscore meanin… repeat 174 174 2.13 173 0.035 0.349 ns
## 10 meanscore meanin… repeat 174 174 2.55 173 0.012 0.117 ns# the post hoc analysis for the main effect of workload dimension
pwc.workload.dimension <- MWdata %>% pairwise_t_test(meanscore ~ dimension, p.adjust.method = "bonferroni", paired = TRUE)
pwc.workload.dimension## # A tibble: 15 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 means… effort frust… 145 145 6.47 144 1.46e-9 2.19e-8 ****
## 2 means… effort mental 145 145 1.13 144 2.6 e-1 1 e+0 ns
## 3 means… effort perfo… 145 145 2.88 144 5 e-3 6.8 e-2 ns
## 4 means… effort physi… 145 145 3.07 144 3 e-3 3.9 e-2 *
## 5 means… effort time 145 145 4.53 144 1.25e-5 1.88e-4 ***
## 6 means… frust… mental 145 145 -3.96 144 1.2 e-4 2 e-3 **
## 7 means… frust… perfo… 145 145 -4.85 144 3.21e-6 4.82e-5 ****
## 8 means… frust… physi… 145 145 -2.86 144 5 e-3 7.3 e-2 ns
## 9 means… frust… time 145 145 -3.39 144 8.89e-4 1.3 e-2 *
## 10 means… mental perfo… 145 145 1.41 144 1.62e-1 1 e+0 ns
## 11 means… mental physi… 145 145 1.63 144 1.05e-1 1 e+0 ns
## 12 means… mental time 145 145 2.04 144 4.3 e-2 6.49e-1 ns
## 13 means… perfo… physi… 145 145 0.190 144 8.49e-1 1 e+0 ns
## 14 means… perfo… time 145 145 0.495 144 6.21e-1 1 e+0 ns
## 15 means… physi… time 145 145 0.227 144 8.21e-1 1 e+0 ns# the simple simple effect analysis for the interaction effect of age group, countermeasure type, and workload dimension
## younger drivers (condition X dimension)
filter.age.Y <- filter(MWdata,agegroup2 =="Y")
### fit model
Model <- lmer(data = filter.age.Y, meanscore~condition*dimension+(1|sub))
summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: meanscore ~ condition * dimension + (1 | sub)
## Data: filter.age.Y
##
## REML criterion at convergence: 4003.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.86685 -0.74177 -0.04496 0.71066 2.13181
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 39.58 6.292
## Residual 642.60 25.350
## Number of obs: 450, groups: sub, 15
##
## Fixed effects:
## Estimate Std. Error df t value
## (Intercept) 46.000 6.744 382.639 6.821
## conditioncontrol 4.333 9.256 406.000 0.468
## conditionmeaningfully 2.333 9.256 406.000 0.252
## conditionmeaninglessly 2.667 9.256 406.000 0.288
## conditionrepeat -13.000 9.256 406.000 -1.404
## dimensionfrustration -27.667 9.256 406.000 -2.989
## dimensionmental 12.667 9.256 406.000 1.368
## dimensionperformance -10.667 9.256 406.000 -1.152
## dimensionphysical -9.000 9.256 406.000 -0.972
## dimensiontime -8.000 9.256 406.000 -0.864
## conditioncontrol:dimensionfrustration 18.000 13.090 406.000 1.375
## conditionmeaningfully:dimensionfrustration 24.333 13.090 406.000 1.859
## conditionmeaninglessly:dimensionfrustration 25.667 13.090 406.000 1.961
## conditionrepeat:dimensionfrustration 18.333 13.090 406.000 1.401
## conditioncontrol:dimensionmental -26.333 13.090 406.000 -2.012
## conditionmeaningfully:dimensionmental -27.667 13.090 406.000 -2.114
## conditionmeaninglessly:dimensionmental -32.000 13.090 406.000 -2.445
## conditionrepeat:dimensionmental 1.000 13.090 406.000 0.076
## conditioncontrol:dimensionperformance 9.333 13.090 406.000 0.713
## conditionmeaningfully:dimensionperformance 15.333 13.090 406.000 1.171
## conditionmeaninglessly:dimensionperformance 11.333 13.090 406.000 0.866
## conditionrepeat:dimensionperformance 8.667 13.090 406.000 0.662
## conditioncontrol:dimensionphysical 5.000 13.090 406.000 0.382
## conditionmeaningfully:dimensionphysical -6.333 13.090 406.000 -0.484
## conditionmeaninglessly:dimensionphysical -1.000 13.090 406.000 -0.076
## conditionrepeat:dimensionphysical 12.667 13.090 406.000 0.968
## conditioncontrol:dimensiontime -2.333 13.090 406.000 -0.178
## conditionmeaningfully:dimensiontime 4.333 13.090 406.000 0.331
## conditionmeaninglessly:dimensiontime 3.667 13.090 406.000 0.280
## conditionrepeat:dimensiontime 2.667 13.090 406.000 0.204
## Pr(>|t|)
## (Intercept) 3.54e-11 ***
## conditioncontrol 0.63993
## conditionmeaningfully 0.80111
## conditionmeaninglessly 0.77342
## conditionrepeat 0.16095
## dimensionfrustration 0.00297 **
## dimensionmental 0.17193
## dimensionperformance 0.24985
## dimensionphysical 0.33148
## dimensiontime 0.38795
## conditioncontrol:dimensionfrustration 0.16987
## conditionmeaningfully:dimensionfrustration 0.06377 .
## conditionmeaninglessly:dimensionfrustration 0.05060 .
## conditionrepeat:dimensionfrustration 0.16212
## conditioncontrol:dimensionmental 0.04492 *
## conditionmeaningfully:dimensionmental 0.03517 *
## conditionmeaninglessly:dimensionmental 0.01493 *
## conditionrepeat:dimensionmental 0.93915
## conditioncontrol:dimensionperformance 0.47626
## conditionmeaningfully:dimensionperformance 0.24215
## conditionmeaninglessly:dimensionperformance 0.38713
## conditionrepeat:dimensionperformance 0.50831
## conditioncontrol:dimensionphysical 0.70269
## conditionmeaningfully:dimensionphysical 0.62878
## conditionmeaninglessly:dimensionphysical 0.93915
## conditionrepeat:dimensionphysical 0.33381
## conditioncontrol:dimensiontime 0.85862
## conditionmeaningfully:dimensiontime 0.74079
## conditionmeaninglessly:dimensiontime 0.77954
## conditionrepeat:dimensiontime 0.83868
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Correlation matrix not shown by default, as p = 30 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need itanova(Model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 7134.8 1783.7 4 406 2.7757 0.02678 *
## dimension 5308.4 1061.7 5 406 1.6522 0.14513
## condition:dimension 24296.6 1214.8 20 406 1.8905 0.01189 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1### pairwise comparisons of five countermeasure types under different workload dimensions
#### (1) mental
filter1.MWdata<- filter(filter.age.Y, dimension =="mental")
library(lmerTest)
Model <- lmer(data = filter1.MWdata, meanscore~condition+(1|sub))
summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: meanscore ~ condition + (1 | sub)
## Data: filter1.MWdata
##
## REML criterion at convergence: 667
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.67994 -0.65291 -0.08546 0.59523 2.26289
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 202.5 14.23
## Residual 536.3 23.16
## Number of obs: 75, groups: sub, 15
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 58.667 7.018 53.823 8.359 2.67e-11 ***
## conditioncontrol -22.000 8.456 56.000 -2.602 0.01185 *
## conditionmeaningfully -25.333 8.456 56.000 -2.996 0.00407 **
## conditionmeaninglessly -29.333 8.456 56.000 -3.469 0.00101 **
## conditionrepeat -12.000 8.456 56.000 -1.419 0.16141
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.602
## cndtnmnngfl -0.602 0.500
## cndtnmnngls -0.602 0.500 0.500
## conditinrpt -0.602 0.500 0.500 0.500anova(Model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 8368 2092 4 56 3.9009 0.007285 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1pwc <- filter1.MWdata %>% pairwise_t_test(meanscore ~ condition, p.adjust.method = "bonferroni", paired = TRUE)
pwc## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 meanscore answer contr… 15 15 2.11 14 0.053 0.534 ns
## 2 meanscore answer meani… 15 15 3.11 14 0.008 0.078 ns
## 3 meanscore answer meani… 15 15 2.90 14 0.012 0.117 ns
## 4 meanscore answer repeat 15 15 1.44 14 0.173 1 ns
## 5 meanscore control meani… 15 15 0.353 14 0.729 1 ns
## 6 meanscore control meani… 15 15 0.791 14 0.442 1 ns
## 7 meanscore control repeat 15 15 -1.36 14 0.195 1 ns
## 8 meanscore meanin… meani… 15 15 0.574 14 0.575 1 ns
## 9 meanscore meanin… repeat 15 15 -2.53 14 0.024 0.24 ns
## 10 meanscore meanin… repeat 15 15 -2.20 14 0.045 0.448 ns#### (2) physical
filter2.MWdata<- filter(filter.age.Y, dimension =="physical")
Model <- lmer(data = filter2.MWdata, meanscore~condition+(1|sub))
summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: meanscore ~ condition + (1 | sub)
## Data: filter2.MWdata
##
## REML criterion at convergence: 658.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.41484 -0.55039 0.00662 0.64018 1.86472
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 379.9 19.49
## Residual 418.8 20.46
## Number of obs: 75, groups: sub, 15
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 37.0000 7.2971 36.7463 5.071 1.15e-05 ***
## conditioncontrol 9.3333 7.4727 56.0000 1.249 0.217
## conditionmeaningfully -4.0000 7.4727 56.0000 -0.535 0.595
## conditionmeaninglessly 1.6667 7.4727 56.0000 0.223 0.824
## conditionrepeat -0.3333 7.4727 56.0000 -0.045 0.965
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.512
## cndtnmnngfl -0.512 0.500
## cndtnmnngls -0.512 0.500 0.500
## conditinrpt -0.512 0.500 0.500 0.500anova(Model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 1456.7 364.17 4 56 0.8695 0.488pwc <- filter2.MWdata %>% pairwise_t_test(meanscore ~ condition, p.adjust.method = "bonferroni", paired = TRUE)
pwc## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 meanscore answer contr… 15 15 -1.66 14 0.119 1 ns
## 2 meanscore answer meani… 15 15 0.616 14 0.548 1 ns
## 3 meanscore answer meani… 15 15 -0.254 14 0.803 1 ns
## 4 meanscore answer repeat 15 15 0.0461 14 0.964 1 ns
## 5 meanscore control meani… 15 15 1.91 14 0.076 0.762 ns
## 6 meanscore control meani… 15 15 1.02 14 0.324 1 ns
## 7 meanscore control repeat 15 15 1.16 14 0.264 1 ns
## 8 meanscore meanin… meani… 15 15 -0.918 14 0.374 1 ns
## 9 meanscore meanin… repeat 15 15 -0.380 14 0.71 1 ns
## 10 meanscore meanin… repeat 15 15 0.218 14 0.831 1 ns#### (3) time
filter3.MWdata<- filter(filter.age.Y, dimension =="time")
Model <- lmer(data = filter3.MWdata, meanscore~condition+(1|sub))## boundary (singular) fit: see help('isSingular')summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: meanscore ~ condition + (1 | sub)
## Data: filter3.MWdata
##
## REML criterion at convergence: 651.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.9257 -0.8398 -0.1158 0.8832 1.8388
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0 0.00
## Residual 530 23.02
## Number of obs: 75, groups: sub, 15
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 38.000 5.944 70.000 6.393 1.57e-08 ***
## conditioncontrol 2.000 8.406 70.000 0.238 0.813
## conditionmeaningfully 6.667 8.406 70.000 0.793 0.430
## conditionmeaninglessly 6.333 8.406 70.000 0.753 0.454
## conditionrepeat -10.333 8.406 70.000 -1.229 0.223
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.707
## cndtnmnngfl -0.707 0.500
## cndtnmnngls -0.707 0.500 0.500
## conditinrpt -0.707 0.500 0.500 0.500
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')anova(Model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 2864.7 716.17 4 70 1.3513 0.2598pwc <- filter3.MWdata %>% pairwise_t_test(meanscore ~ condition, p.adjust.method = "bonferroni", paired = TRUE)
pwc## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 meanscore answer contr… 15 15 -0.276 14 0.787 1 ns
## 2 meanscore answer meani… 15 15 -0.974 14 0.346 1 ns
## 3 meanscore answer meani… 15 15 -0.616 14 0.548 1 ns
## 4 meanscore answer repeat 15 15 1.73 14 0.105 1 ns
## 5 meanscore control meani… 15 15 -0.538 14 0.599 1 ns
## 6 meanscore control meani… 15 15 -0.409 14 0.689 1 ns
## 7 meanscore control repeat 15 15 1.56 14 0.141 1 ns
## 8 meanscore meanin… meani… 15 15 0.0325 14 0.975 1 ns
## 9 meanscore meanin… repeat 15 15 2.46 14 0.028 0.276 ns
## 10 meanscore meanin… repeat 15 15 1.77 14 0.098 0.982 ns#### (4) effort
filter4.MWdata<- filter(filter.age.Y, dimension =="effort")
Model <- lmer(data = filter4.MWdata, meanscore~condition+(1|sub))## boundary (singular) fit: see help('isSingular')summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: meanscore ~ condition + (1 | sub)
## Data: filter4.MWdata
##
## REML criterion at convergence: 675
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.6009 -0.9288 0.2322 0.7088 1.8209
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 8.045e-13 8.970e-07
## Residual 7.440e+02 2.728e+01
## Number of obs: 75, groups: sub, 15
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 46.000 7.043 70.000 6.532 8.82e-09 ***
## conditioncontrol 4.333 9.960 70.000 0.435 0.665
## conditionmeaningfully 2.333 9.960 70.000 0.234 0.815
## conditionmeaninglessly 2.667 9.960 70.000 0.268 0.790
## conditionrepeat -13.000 9.960 70.000 -1.305 0.196
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.707
## cndtnmnngfl -0.707 0.500
## cndtnmnngls -0.707 0.500 0.500
## conditinrpt -0.707 0.500 0.500 0.500
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')anova(Model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 2964.7 741.17 4 70 0.9962 0.4155pwc <- filter4.MWdata %>% pairwise_t_test(meanscore ~ condition, p.adjust.method = "bonferroni", paired = TRUE)
pwc## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 meanscore answer contr… 15 15 -0.496 14 0.628 1 ns
## 2 meanscore answer meani… 15 15 -0.300 14 0.768 1 ns
## 3 meanscore answer meani… 15 15 -0.226 14 0.825 1 ns
## 4 meanscore answer repeat 15 15 1.67 14 0.117 1 ns
## 5 meanscore control meani… 15 15 0.207 14 0.839 1 ns
## 6 meanscore control meani… 15 15 0.136 14 0.894 1 ns
## 7 meanscore control repeat 15 15 2.36 14 0.033 0.334 ns
## 8 meanscore meanin… meani… 15 15 -0.0289 14 0.977 1 ns
## 9 meanscore meanin… repeat 15 15 1.55 14 0.143 1 ns
## 10 meanscore meanin… repeat 15 15 1.33 14 0.203 1 ns#### (5) frustration
filter5.MWdata<- filter(filter.age.Y, dimension =="frustration")
Model <- lmer(data = filter5.MWdata, meanscore~condition+(1|sub))
summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: meanscore ~ condition + (1 | sub)
## Data: filter5.MWdata
##
## REML criterion at convergence: 665.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.7843 -0.7922 -0.0414 0.5754 1.9991
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 61.11 7.817
## Residual 600.80 24.511
## Number of obs: 75, groups: sub, 15
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 18.333 6.643 67.692 2.760 0.00743 **
## conditioncontrol 22.333 8.950 56.000 2.495 0.01556 *
## conditionmeaningfully 26.667 8.950 56.000 2.979 0.00426 **
## conditionmeaninglessly 28.333 8.950 56.000 3.166 0.00250 **
## conditionrepeat 5.333 8.950 56.000 0.596 0.55365
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.674
## cndtnmnngfl -0.674 0.500
## cndtnmnngls -0.674 0.500 0.500
## conditinrpt -0.674 0.500 0.500 0.500anova(Model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 10115 2528.8 4 56 4.2091 0.004753 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1pwc <- filter5.MWdata %>% pairwise_t_test(meanscore ~ condition, p.adjust.method = "bonferroni", paired = TRUE)
pwc## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 meanscore answer contr… 15 15 -2.96 14 0.01 0.104 ns
## 2 meanscore answer meani… 15 15 -2.91 14 0.011 0.113 ns
## 3 meanscore answer meani… 15 15 -3.16 14 0.007 0.069 ns
## 4 meanscore answer repeat 15 15 -0.764 14 0.457 1 ns
## 5 meanscore control meani… 15 15 -0.541 14 0.597 1 ns
## 6 meanscore control meani… 15 15 -0.645 14 0.529 1 ns
## 7 meanscore control repeat 15 15 2.39 14 0.032 0.316 ns
## 8 meanscore meanin… meani… 15 15 -0.156 14 0.878 1 ns
## 9 meanscore meanin… repeat 15 15 1.99 14 0.066 0.664 ns
## 10 meanscore meanin… repeat 15 15 2.28 14 0.039 0.387 ns#### (6) performance
filter6.MWdata<- filter(filter.age.Y, dimension =="performance")
Model <- lmer(data = filter6.MWdata, meanscore~condition+(1|sub))## boundary (singular) fit: see help('isSingular')summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: meanscore ~ condition + (1 | sub)
## Data: filter6.MWdata
##
## REML criterion at convergence: 662.2
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.78095 -0.68292 0.02678 0.76326 1.96842
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 0.0 0.00
## Residual 619.7 24.89
## Number of obs: 75, groups: sub, 15
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 35.333 6.427 70.000 5.497 5.89e-07 ***
## conditioncontrol 13.667 9.090 70.000 1.504 0.137
## conditionmeaningfully 17.667 9.090 70.000 1.944 0.056 .
## conditionmeaninglessly 14.000 9.090 70.000 1.540 0.128
## conditionrepeat -4.333 9.090 70.000 -0.477 0.635
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.707
## cndtnmnngfl -0.707 0.500
## cndtnmnngls -0.707 0.500 0.500
## conditinrpt -0.707 0.500 0.500 0.500
## optimizer (nloptwrap) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')anova(Model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 5662 1415.5 4 70 2.2843 0.06881 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1pwc <- filter6.MWdata %>% pairwise_t_test(meanscore ~ condition, p.adjust.method = "bonferroni", paired = TRUE)
pwc## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 meanscore answer contr… 15 15 -1.39 14 0.187 1 ns
## 2 meanscore answer meani… 15 15 -1.66 14 0.119 1 ns
## 3 meanscore answer meani… 15 15 -1.17 14 0.261 1 ns
## 4 meanscore answer repeat 15 15 0.573 14 0.575 1 ns
## 5 meanscore control meani… 15 15 -0.475 14 0.642 1 ns
## 6 meanscore control meani… 15 15 -0.0387 14 0.97 1 ns
## 7 meanscore control repeat 15 15 1.94 14 0.073 0.729 ns
## 8 meanscore meanin… meani… 15 15 0.458 14 0.654 1 ns
## 9 meanscore meanin… repeat 15 15 2.04 14 0.06 0.605 ns
## 10 meanscore meanin… repeat 15 15 1.80 14 0.093 0.931 ns## middle-aged drivers (condition X dimension)
filter.age.M <- filter(MWdata,agegroup2 =="M")
### fit model
Model <- lmer(data = filter.age.M, meanscore~condition*dimension+(1|sub))
summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: meanscore ~ condition * dimension + (1 | sub)
## Data: filter.age.M
##
## REML criterion at convergence: 3594.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.36082 -0.64771 -0.01516 0.67881 2.54687
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 172.9 13.15
## Residual 441.8 21.02
## Number of obs: 420, groups: sub, 14
##
## Fixed effects:
## Estimate Std. Error df
## (Intercept) 52.8571 6.6259 118.3993
## conditioncontrol 0.7143 7.9442 377.0000
## conditionmeaningfully -8.5714 7.9442 377.0000
## conditionmeaninglessly 7.8571 7.9442 377.0000
## conditionrepeat -1.7033 7.9442 377.0000
## dimensionfrustration -19.2857 7.9442 377.0000
## dimensionmental -0.7143 7.9442 377.0000
## dimensionperformance -16.0714 7.9442 377.0000
## dimensionphysical -10.0000 7.9442 377.0000
## dimensiontime -5.7143 7.9442 377.0000
## conditioncontrol:dimensionfrustration 2.8571 11.2348 377.0000
## conditionmeaningfully:dimensionfrustration 3.9286 11.2348 377.0000
## conditionmeaninglessly:dimensionfrustration -4.6429 11.2348 377.0000
## conditionrepeat:dimensionfrustration -0.7143 11.2348 377.0000
## conditioncontrol:dimensionmental -0.7143 11.2348 377.0000
## conditionmeaningfully:dimensionmental 17.1429 11.2348 377.0000
## conditionmeaninglessly:dimensionmental -13.2143 11.2348 377.0000
## conditionrepeat:dimensionmental -3.5165 11.2348 377.0000
## conditioncontrol:dimensionperformance 6.4286 11.2348 377.0000
## conditionmeaningfully:dimensionperformance 9.6429 11.2348 377.0000
## conditionmeaninglessly:dimensionperformance -1.4286 11.2348 377.0000
## conditionrepeat:dimensionperformance 3.7637 11.2348 377.0000
## conditioncontrol:dimensionphysical 6.0714 11.2348 377.0000
## conditionmeaningfully:dimensionphysical 15.0000 11.2348 377.0000
## conditionmeaninglessly:dimensionphysical -6.7857 11.2348 377.0000
## conditionrepeat:dimensionphysical -3.0769 11.2348 377.0000
## conditioncontrol:dimensiontime -8.9286 11.2348 377.0000
## conditionmeaningfully:dimensiontime -2.5000 11.2348 377.0000
## conditionmeaninglessly:dimensiontime -6.7857 11.2348 377.0000
## conditionrepeat:dimensiontime -0.4396 11.2348 377.0000
## t value Pr(>|t|)
## (Intercept) 7.977 1.06e-12 ***
## conditioncontrol 0.090 0.9284
## conditionmeaningfully -1.079 0.2813
## conditionmeaninglessly 0.989 0.3233
## conditionrepeat -0.214 0.8303
## dimensionfrustration -2.428 0.0157 *
## dimensionmental -0.090 0.9284
## dimensionperformance -2.023 0.0438 *
## dimensionphysical -1.259 0.2089
## dimensiontime -0.719 0.4724
## conditioncontrol:dimensionfrustration 0.254 0.7994
## conditionmeaningfully:dimensionfrustration 0.350 0.7268
## conditionmeaninglessly:dimensionfrustration -0.413 0.6797
## conditionrepeat:dimensionfrustration -0.064 0.9493
## conditioncontrol:dimensionmental -0.064 0.9493
## conditionmeaningfully:dimensionmental 1.526 0.1279
## conditionmeaninglessly:dimensionmental -1.176 0.2403
## conditionrepeat:dimensionmental -0.313 0.7545
## conditioncontrol:dimensionperformance 0.572 0.5675
## conditionmeaningfully:dimensionperformance 0.858 0.3913
## conditionmeaninglessly:dimensionperformance -0.127 0.8989
## conditionrepeat:dimensionperformance 0.335 0.7378
## conditioncontrol:dimensionphysical 0.540 0.5892
## conditionmeaningfully:dimensionphysical 1.335 0.1826
## conditionmeaninglessly:dimensionphysical -0.604 0.5462
## conditionrepeat:dimensionphysical -0.274 0.7843
## conditioncontrol:dimensiontime -0.795 0.4273
## conditionmeaningfully:dimensiontime -0.223 0.8240
## conditionmeaninglessly:dimensiontime -0.604 0.5462
## conditionrepeat:dimensiontime -0.039 0.9688
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Correlation matrix not shown by default, as p = 30 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need itanova(Model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 1336.0 334.0 4 377 0.7561 0.5545
## dimension 18081.9 3616.4 5 377 8.1860 2.364e-07 ***
## condition:dimension 6577.7 328.9 20 377 0.7445 0.7793
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1### pairwise comparisons of five countermeasure types under different workload dimensions
#### (1) mental
filter1.MWdata<- filter(filter.age.M, dimension =="mental")
Model <- lmer(data = filter1.MWdata, meanscore~condition+(1|sub))
summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: meanscore ~ condition + (1 | sub)
## Data: filter1.MWdata
##
## REML criterion at convergence: 630.9
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.86611 -0.77014 0.00712 0.73527 1.70739
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 153.8 12.40
## Residual 674.2 25.97
## Number of obs: 70, groups: sub, 14
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 5.214e+01 7.691e+00 5.712e+01 6.780 7.33e-09 ***
## conditioncontrol -1.340e-14 9.814e+00 5.200e+01 0.000 1.000
## conditionmeaningfully 8.571e+00 9.814e+00 5.200e+01 0.873 0.386
## conditionmeaninglessly -5.357e+00 9.814e+00 5.200e+01 -0.546 0.587
## conditionrepeat -5.220e+00 9.814e+00 5.200e+01 -0.532 0.597
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.638
## cndtnmnngfl -0.638 0.500
## cndtnmnngls -0.638 0.500 0.500
## conditinrpt -0.638 0.500 0.500 0.500anova(Model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 1800.5 450.14 4 52 0.6676 0.6173pwc <- filter1.MWdata %>% pairwise_t_test(meanscore ~ condition, p.adjust.method = "bonferroni", paired = TRUE)
pwc## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 meanscore answer contr… 14 14 0 13 1 1 ns
## 2 meanscore answer meani… 14 14 -1.05 13 0.314 1 ns
## 3 meanscore answer meani… 14 14 0.467 13 0.648 1 ns
## 4 meanscore answer repeat 14 14 0.472 13 0.645 1 ns
## 5 meanscore control meani… 14 14 -1.16 13 0.265 1 ns
## 6 meanscore control meani… 14 14 0.648 13 0.528 1 ns
## 7 meanscore control repeat 14 14 0.462 13 0.652 1 ns
## 8 meanscore meanin… meani… 14 14 1.66 13 0.121 1 ns
## 9 meanscore meanin… repeat 14 14 1.21 13 0.249 1 ns
## 10 meanscore meanin… repeat 14 14 -0.0114 13 0.991 1 ns#### (2) physical
filter2.MWdata<- filter(filter.age.M, dimension =="physical")
Model <- lmer(data = filter2.MWdata, meanscore~condition+(1|sub))
summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: meanscore ~ condition + (1 | sub)
## Data: filter2.MWdata
##
## REML criterion at convergence: 608
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.3478 -0.5353 -0.0379 0.6914 1.8982
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 321.0 17.92
## Residual 399.6 19.99
## Number of obs: 70, groups: sub, 14
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 42.857 7.174 36.239 5.974 7.35e-07 ***
## conditioncontrol 6.786 7.556 52.000 0.898 0.373
## conditionmeaningfully 6.429 7.556 52.000 0.851 0.399
## conditionmeaninglessly 1.071 7.556 52.000 0.142 0.888
## conditionrepeat -4.780 7.556 52.000 -0.633 0.530
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.527
## cndtnmnngfl -0.527 0.500
## cndtnmnngls -0.527 0.500 0.500
## conditinrpt -0.527 0.500 0.500 0.500anova(Model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 1306.2 326.55 4 52 0.8172 0.5201pwc <- filter2.MWdata %>% pairwise_t_test(meanscore ~ condition, p.adjust.method = "bonferroni", paired = TRUE)
pwc## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 meanscore answer contr… 14 14 -0.842 13 0.415 1 ns
## 2 meanscore answer meani… 14 14 -0.956 13 0.356 1 ns
## 3 meanscore answer meani… 14 14 -0.117 13 0.908 1 ns
## 4 meanscore answer repeat 14 14 0.783 13 0.448 1 ns
## 5 meanscore control meani… 14 14 0.0660 13 0.948 1 ns
## 6 meanscore control meani… 14 14 0.688 13 0.503 1 ns
## 7 meanscore control repeat 14 14 1.79 13 0.097 0.967 ns
## 8 meanscore meanin… meani… 14 14 0.660 13 0.52 1 ns
## 9 meanscore meanin… repeat 14 14 1.47 13 0.165 1 ns
## 10 meanscore meanin… repeat 14 14 0.670 13 0.514 1 ns#### (3) time
filter3.MWdata<- filter(filter.age.M, dimension =="time")
Model <- lmer(data = filter3.MWdata, meanscore~condition+(1|sub))
summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: meanscore ~ condition + (1 | sub)
## Data: filter3.MWdata
##
## REML criterion at convergence: 580
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -1.93772 -0.59329 0.05874 0.63833 1.93671
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 135.3 11.63
## Residual 280.6 16.75
## Number of obs: 70, groups: sub, 14
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 47.143 5.450 45.659 8.649 3.54e-11 ***
## conditioncontrol -8.214 6.331 52.000 -1.297 0.2002
## conditionmeaningfully -11.071 6.331 52.000 -1.749 0.0862 .
## conditionmeaninglessly 1.071 6.331 52.000 0.169 0.8663
## conditionrepeat -2.143 6.331 52.000 -0.338 0.7364
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.581
## cndtnmnngfl -0.581 0.500
## cndtnmnngls -0.581 0.500 0.500
## conditinrpt -0.581 0.500 0.500 0.500anova(Model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 1580.7 395.18 4 52 1.4085 0.2441pwc <- filter3.MWdata %>% pairwise_t_test(meanscore ~ condition, p.adjust.method = "bonferroni", paired = TRUE)
pwc## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 meanscore answer contr… 14 14 1.24 13 0.235 1 ns
## 2 meanscore answer meani… 14 14 1.62 13 0.128 1 ns
## 3 meanscore answer meani… 14 14 -0.140 13 0.891 1 ns
## 4 meanscore answer repeat 14 14 0.442 13 0.666 1 ns
## 5 meanscore control meani… 14 14 0.528 13 0.607 1 ns
## 6 meanscore control meani… 14 14 -1.43 13 0.177 1 ns
## 7 meanscore control repeat 14 14 -1.15 13 0.272 1 ns
## 8 meanscore meanin… meani… 14 14 -1.82 13 0.092 0.925 ns
## 9 meanscore meanin… repeat 14 14 -1.56 13 0.143 1 ns
## 10 meanscore meanin… repeat 14 14 0.449 13 0.661 1 ns#### (4) effort
filter4.MWdata<- filter(filter.age.M, dimension =="effort")
Model <- lmer(data = filter4.MWdata, meanscore~condition+(1|sub))
summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: meanscore ~ condition + (1 | sub)
## Data: filter4.MWdata
##
## REML criterion at convergence: 604.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.00403 -0.74826 0.07516 0.60586 1.72410
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 186.0 13.64
## Residual 413.3 20.33
## Number of obs: 70, groups: sub, 14
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 52.8571 6.5424 46.9252 8.079 1.96e-10 ***
## conditioncontrol 0.7143 7.6838 52.0000 0.093 0.926
## conditionmeaningfully -8.5714 7.6838 52.0000 -1.116 0.270
## conditionmeaninglessly 7.8571 7.6838 52.0000 1.023 0.311
## conditionrepeat -1.7033 7.6838 52.0000 -0.222 0.825
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.587
## cndtnmnngfl -0.587 0.500
## cndtnmnngls -0.587 0.500 0.500
## conditinrpt -0.587 0.500 0.500 0.500anova(Model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 1932.5 483.12 4 52 1.169 0.3352pwc <- filter4.MWdata %>% pairwise_t_test(meanscore ~ condition, p.adjust.method = "bonferroni", paired = TRUE)
pwc## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 meanscore answer contr… 14 14 -0.122 13 0.905 1 ns
## 2 meanscore answer meani… 14 14 1.27 13 0.226 1 ns
## 3 meanscore answer meani… 14 14 -0.878 13 0.396 1 ns
## 4 meanscore answer repeat 14 14 0.247 13 0.809 1 ns
## 5 meanscore control meani… 14 14 1.71 13 0.11 1 ns
## 6 meanscore control meani… 14 14 -0.790 13 0.444 1 ns
## 7 meanscore control repeat 14 14 0.317 13 0.756 1 ns
## 8 meanscore meanin… meani… 14 14 -2.36 13 0.035 0.346 ns
## 9 meanscore meanin… repeat 14 14 -0.845 13 0.413 1 ns
## 10 meanscore meanin… repeat 14 14 0.960 13 0.355 1 ns#### (5) frustration
filter5.MWdata<- filter(filter.age.M, dimension =="frustration")
Model <- lmer(data = filter5.MWdata, meanscore~condition+(1|sub))
summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: meanscore ~ condition + (1 | sub)
## Data: filter5.MWdata
##
## REML criterion at convergence: 593.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.0913 -0.5418 -0.1313 0.4585 3.0320
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 156.3 12.50
## Residual 348.4 18.67
## Number of obs: 70, groups: sub, 14
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 33.571 6.004 46.985 5.591 1.11e-06 ***
## conditioncontrol 3.571 7.055 52.000 0.506 0.615
## conditionmeaningfully -4.643 7.055 52.000 -0.658 0.513
## conditionmeaninglessly 3.214 7.055 52.000 0.456 0.651
## conditionrepeat -2.418 7.055 52.000 -0.343 0.733
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.588
## cndtnmnngfl -0.588 0.500
## cndtnmnngls -0.588 0.500 0.500
## conditinrpt -0.588 0.500 0.500 0.500anova(Model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 706.61 176.65 4 52 0.507 0.7308pwc <- filter5.MWdata %>% pairwise_t_test(meanscore ~ condition, p.adjust.method = "bonferroni", paired = TRUE)
pwc## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 meanscore answer contr… 14 14 -0.608 13 0.553 1 ns
## 2 meanscore answer meani… 14 14 0.687 13 0.504 1 ns
## 3 meanscore answer meani… 14 14 -0.410 13 0.688 1 ns
## 4 meanscore answer repeat 14 14 0.435 13 0.67 1 ns
## 5 meanscore control meani… 14 14 1.49 13 0.16 1 ns
## 6 meanscore control meani… 14 14 0.0577 13 0.955 1 ns
## 7 meanscore control repeat 14 14 0.821 13 0.426 1 ns
## 8 meanscore meanin… meani… 14 14 -1.37 13 0.194 1 ns
## 9 meanscore meanin… repeat 14 14 -0.269 13 0.792 1 ns
## 10 meanscore meanin… repeat 14 14 0.557 13 0.587 1 ns#### (6) performance
filter6.MWdata<- filter(filter.age.M, dimension =="performance")
Model <- lmer(data = filter6.MWdata, meanscore~condition+(1|sub))
summary(Model)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: meanscore ~ condition + (1 | sub)
## Data: filter6.MWdata
##
## REML criterion at convergence: 601.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.0122 -0.4883 -0.1755 0.5333 3.1298
##
## Random effects:
## Groups Name Variance Std.Dev.
## sub (Intercept) 246.0 15.68
## Residual 373.4 19.32
## Number of obs: 70, groups: sub, 14
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 36.786 6.652 39.858 5.530 2.18e-06 ***
## conditioncontrol 7.143 7.304 52.000 0.978 0.333
## conditionmeaningfully 1.071 7.304 52.000 0.147 0.884
## conditionmeaninglessly 6.429 7.304 52.000 0.880 0.383
## conditionrepeat 2.060 7.304 52.000 0.282 0.779
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) cndtnc cndtnmnngf cndtnmnngl
## condtncntrl -0.549
## cndtnmnngfl -0.549 0.500
## cndtnmnngls -0.549 0.500 0.500
## conditinrpt -0.549 0.500 0.500 0.500anova(Model)## Type III Analysis of Variance Table with Satterthwaite's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## condition 587.16 146.79 4 52 0.3931 0.8127pwc <- filter6.MWdata %>% pairwise_t_test(meanscore ~ condition, p.adjust.method = "bonferroni", paired = TRUE)
pwc## # A tibble: 10 × 10
## .y. group1 group2 n1 n2 statistic df p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <dbl> <chr>
## 1 meanscore answer contr… 14 14 -1.37 13 0.193 1 ns
## 2 meanscore answer meani… 14 14 -0.275 13 0.787 1 ns
## 3 meanscore answer meani… 14 14 -0.791 13 0.443 1 ns
## 4 meanscore answer repeat 14 14 -0.301 13 0.769 1 ns
## 5 meanscore control meani… 14 14 0.968 13 0.351 1 ns
## 6 meanscore control meani… 14 14 0.117 13 0.909 1 ns
## 7 meanscore control repeat 14 14 0.658 13 0.522 1 ns
## 8 meanscore meanin… meani… 14 14 -0.654 13 0.525 1 ns
## 9 meanscore meanin… repeat 14 14 -0.138 13 0.892 1 ns
## 10 meanscore meanin… repeat 14 14 0.394 13 0.7 1 ns3.6.2 Acceptance
# read data
USdata <- read_excel("/Users/betty/Desktop/Acceptance (processed).xlsx", sheet=1)
# as factors & groupby
USdata$condition <- as.factor(USdata$condition)
USdata$agegroup2 <- as.factor(USdata$agegroup2)
USdata %>%
group_by(agegroup2,condition)## # A tibble: 116 × 7
## # Groups: agegroup2, condition [8]
## sub age agegroup2 order condition usefulness satisfaction
## <chr> <dbl> <fct> <dbl> <fct> <dbl> <dbl>
## 1 sub1 42 M 2 meaninglessly -1.6 -0.25
## 2 sub1 42 M 4 meaningfully 0.8 0.25
## 3 sub1 42 M 5 repeat 0.6 0.25
## 4 sub1 42 M 3 answer 1.4 0.25
## 5 sub10 36 M 4 meaninglessly -0.2 -0.5
## 6 sub10 36 M 3 meaningfully -1.8 -1.75
## 7 sub10 36 M 1 repeat -1.6 -1.75
## 8 sub10 36 M 2 answer -0.4 -0.5
## 9 sub11 28 Y 4 meaninglessly -1.6 -0.5
## 10 sub11 28 Y 3 meaningfully -0.8 0
## # ℹ 106 more rows# Analysis of younger drivers
filter.Y<- filter(USdata,agegroup2 =="Y")
## main effect of countermeasure type on usefulness
Us.fried.useful.young <- filter.Y %>% friedman_test(usefulness ~ condition |sub)
Us.fried.useful.young## # A tibble: 1 × 6
## .y. n statistic df p method
## * <chr> <int> <dbl> <dbl> <dbl> <chr>
## 1 usefulness 15 24.1 3 0.0000233 Friedman test## the post hoc analysis for the main effect of countermeasure type on usefulness
pwc.useful.young <- filter.Y %>% wilcox_test(usefulness ~ condition, paired = TRUE, p.adjust.method = "bonferroni")
pwc.useful.young## # A tibble: 6 × 9
## .y. group1 group2 n1 n2 statistic p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <chr>
## 1 usefulness answer meani… 15 15 115 2 e-3 0.012 *
## 2 usefulness answer meani… 15 15 119 8.9e-4 0.005 **
## 3 usefulness answer repeat 15 15 53.5 7.4e-2 0.442 ns
## 4 usefulness meaningfully meani… 15 15 82.5 6.4e-2 0.382 ns
## 5 usefulness meaningfully repeat 15 15 15 3.6e-2 0.215 ns
## 6 usefulness meaningless… repeat 15 15 10 5 e-3 0.029 *## main effect of countermeasure type on satisfaction
Us.fried.satisfaction.young <- filter.Y %>% friedman_test(satisfaction ~ condition |sub)
Us.fried.satisfaction.young## # A tibble: 1 × 6
## .y. n statistic df p method
## * <chr> <int> <dbl> <dbl> <dbl> <chr>
## 1 satisfaction 15 19.5 3 0.000219 Friedman test## the post hoc analysis for the main effect of countermeasure type on satisfaction
pwc.satisfaction.young <- filter.Y %>% wilcox_test(satisfaction ~ condition, paired = TRUE, p.adjust.method = "bonferroni")
pwc.satisfaction.young## # A tibble: 6 × 9
## .y. group1 group2 n1 n2 statistic p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <chr>
## 1 satisfaction answer meani… 15 15 94 0.01 0.059 ns
## 2 satisfaction answer meani… 15 15 118 0.001 0.006 **
## 3 satisfaction answer repeat 15 15 44 0.101 0.606 ns
## 4 satisfaction meaningful… meani… 15 15 89.5 0.022 0.13 ns
## 5 satisfaction meaningful… repeat 15 15 25.5 0.17 1 ns
## 6 satisfaction meaningles… repeat 15 15 10.5 0.005 0.032 *# Analysis of middle-aged drivers
filter.M <- filter(USdata,agegroup2 =="M")
## main effect of countermeasure type on usefulness
Us.fried.useful.middle <- filter.M %>% friedman_test(usefulness ~ condition |sub)
Us.fried.useful.middle## # A tibble: 1 × 6
## .y. n statistic df p method
## * <chr> <int> <dbl> <dbl> <dbl> <chr>
## 1 usefulness 14 14.6 3 0.00218 Friedman test## the post hoc analysis for the main effect of countermeasure type on usefulness
pwc.useful.middle <- filter.M %>% wilcox_test(usefulness ~ condition, paired = TRUE, p.adjust.method = "bonferroni")
pwc.useful.middle## # A tibble: 6 × 9
## .y. group1 group2 n1 n2 statistic p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <chr>
## 1 usefulness answer meani… 14 14 67 0.031 0.184 ns
## 2 usefulness answer meani… 14 14 94 0.01 0.06 ns
## 3 usefulness answer repeat 14 14 56.5 0.181 1 ns
## 4 usefulness meaningfully meani… 14 14 82 0.068 0.411 ns
## 5 usefulness meaningfully repeat 14 14 12.5 0.041 0.247 ns
## 6 usefulness meaninglessly repeat 14 14 11 0.017 0.104 ns## main effect of countermeasure type on satisfaction
Us.fried.satisfaction.middle <- filter.M %>% friedman_test(satisfaction ~ condition |sub)
Us.fried.satisfaction.middle## # A tibble: 1 × 6
## .y. n statistic df p method
## * <chr> <int> <dbl> <dbl> <dbl> <chr>
## 1 satisfaction 14 4.36 3 0.225 Friedman test## the post hoc analysis for the main effect of countermeasure type on usefulness
pwc.satisfaction.middle <- filter.M %>% wilcox_test(satisfaction ~ condition, paired = TRUE, p.adjust.method = "bonferroni")
pwc.satisfaction.middle## # A tibble: 6 × 9
## .y. group1 group2 n1 n2 statistic p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <chr>
## 1 satisfaction answer meani… 14 14 60 0.105 0.63 ns
## 2 satisfaction answer meani… 14 14 65 0.045 0.272 ns
## 3 satisfaction answer repeat 14 14 30 0.401 1 ns
## 4 satisfaction meaningful… meani… 14 14 59.5 0.344 1 ns
## 5 satisfaction meaningful… repeat 14 14 14.5 0.103 0.618 ns
## 6 satisfaction meaningles… repeat 14 14 28.5 0.139 0.834 ns